Do y'all think the Space-X Super heavy/Star ship would work out?

[Rakaydos]

35 minutes ago, Starman4308 said:

Independent engines-out are hardly the only failure mode imaginable.

COPV fails, and ignites the tank (which, you know, happened on a Falcon 9).

Control glitch cuts the Superheavy thrust half a second into flight.

The active cooling on the Starship reentry fails, even on a small patch, leading to burnthrough and vessel breakup on reentry.

And that's just known failure modes. Take the Columbia disaster as an example: nobody thought foamstrikes could be a failure mode until a foamstrike was a failure mode.

 

The Starship is experimental technology, in a class known to be prone to failure. It's going to be a huge amount of effort to man-rate it.

1)There's no COPVs in the tanks on Starship.

2)you're going to have to be more specific as to the failure here- do you mean an insufficently debugged control algorithum, which the Water Tower hopper will be testing in a few months? Comunication error that should have caused a hold in the countdown? A multiple engine problem too severe to throttle back and abort to launch clamps?

3) Failure of active cooling to what? oversize the cooling channals so that any one channal can fail and the adjacent ones can cover the load, backed by a heat pipe that can redistribute a more general failure's heat load to the remainder of the funtional shield. The whole system can be tested prior to entry, unlike an ablative shield, and can be ran with leaks long enough to aerocapture if they divert landing propellant to the heat shield.

1 hour ago, Xd the great said:

What about the flip manouvers and the meteoroid shield being the heatshield?

The DC-X managed flip maneuvers with engine gimbal only. RCS is supposed to be in the 10-ton class, so gimbal+RCS should be plenty to make up for an actuator failure. (plus, the fin actuators themselves would be redundant on each fin)

And stainless steel can be welded in space, with enough warning. (if not, well, see previus post about diverting landing methane to the heat shield)

1 hour ago, tater said:

That's one failure mode.

What about a catastrophic RUD event? Why is 1:270 so hard for both SpaceX and Boeing to demonstrate with far simpler spacecraft that land with parachutes (very well understood tech)?

In 2 words: Mass budget. Starship has it, Dragon II and Starliner do not.

[Starman4308]

21 minutes ago, Rakaydos said:

1)There's no COPVs in the tanks on Starship.

2)you're going to have to be more specific as to the failure here- do you mean an insufficently debugged control algorithum, which the Water Tower hopper will be testing in a few months? Comunication error that should have caused a hold in the countdown? A multiple engine problem too severe to throttle back and abort to launch clamps?

3) Failure of active cooling to what? oversize the cooling channals so that any one channal can fail and the adjacent ones can cover the load, backed by a heat pipe that can redistribute a more general failure's heat load to the remainder of the funtional shield. The whole system can be tested prior to entry, unlike an ablative shield, and can be ran with leaks long enough to aerocapture if they divert landing propellant to the heat shield.

1) Technically you have me there. Practically, substitute almost any other system of the vehicle in, and that's still something that could cause a failure.

2) Does it matter? With experimental vehicles, half the time it's not what you anticipate might be an issue that kills you: it's what you don't anticipate.

3) Failure of active cooling to the heatshield during reentry. I don't care how. Maybe a bolt shakes loose on reentry and hits enough adjacent pipes/valves that you get a local hotspot.

The key point here is that rattling off a list of technical details doesn't matter. In vehicles this complex, being run this close to engineering limits, something will be overlooked. Even in the airline industry, overlooked failure modes have caused disasters. The difference with the airline industry is that they've made so many flights that, if there is something overlooked, it's such a strange, rare edge case that very, very few flights will be affected. Orbital launch vehicles? They don't have that track record.

It's not the part you can think of that will fail you. It's the insulation foam falling onto your wing. It's a worker inserting an accelerometer backwards. It's the bellows surrounding the augmented spark injector encountering vibrations in the vacuum of space, that hadn't been an issue on the ground because of condensation (Saturn V Owner's Workshop Manual, pg. 149-150). It's an improperly secured nut that the prior worker used an extra-long wrench to firmly insert, while the new worker used the prescribed (too-short) wrench.

With such massively complicated systems, the only way to get reliable failure-rate data is to go out there and launch some vehicles. Until SpaceX has that, it's criminally risky to put astronauts on something that doesn't have a bone-simple and robust launch escape system.

EDIT: You want to know why Challenger and Columbia killed 14 astronauts? Really, it wasn't the foam, or the SRBs failing in cold weather. It was the poor management decisions combined with an overly experimental vehicle lacking good abort modes that killed 14 astronauts. Fix one problem, and another crops up. I can only hope that SpaceX's actual engineers are a lot more cautious than that.

Edited January 22, 2019 by Starman4308

[Rakaydos]

Starship will not be flying crew until after it's put over ten thousand satelites up, flying unmanned. Yes, before 2023's #DearMoon. Starship is being designed around a flight rate that can practically demonstrate NASA's 1/270 failure rate requirement with actual flights, not just theoretical calculations. The unknown Unknowns will be found through testing, just as the COPV's little Solid Oxygen problem was found and solved before crew flies on Falcon 9.

[Starman4308]

2 minutes ago, Rakaydos said:

Starship will not be flying crew until after it's put over ten thousand satelites up, flying unmanned. Yes, before 2023's #DearMoon. Starship is being designed around a flight rate that can practically demonstrate NASA's 1/270 failure rate requirement with actual flights, not just theoretical calculations. The unknown Unknowns will be found through testing, just as the COPV's little Solid Oxygen problem was found and solved before crew flies on Falcon 9.

I'll believe a 2023 manned launch when I see it.

The timelines Musk predicts are widely known to be aspirational at best and outright pixie dust in many cases. Four years is not much time to go from technology demonstrator to a mature launch system. It took eight years for the Falcon 9 to go from its first flight to a mature configuration, and that was based on much more well-understood design principles.

[Rakaydos]

6 minutes ago, Starman4308 said:

I'll believe a 2023 manned launch when I see it.

The timelines Musk predicts are widely known to be aspirational at best and outright pixie dust in many cases. Four years is not much time to go from technology demonstrator to a mature launch system. It took eight years for the Falcon 9 to go from its first flight to a mature configuration, and that was based on much more well-understood design principles.

8 years from first flight to final upgrade... I'll agree that Starship will be improved on for at least that long. But it will be flying crew before it flies VacRaptors, as per the dearmoon presentation. And the steel based redesign apparently moved their timeline to the right, after affirming the schedule mentioned in 2017.

 

[Starman4308]

1 minute ago, Rakaydos said:

8 years from first flight to final upgrade... I'll agree that Starship will be improved on for at least that long. But it will be flying crew before it flies VacRaptors, as per the dearmoon presentation. And the steel based redesign apparently moved their timeline to the right, after affirming the schedule mentioned in 2017.

 

You do recall the bit where NASA refused to count anything other than Block 5 for validation flights for commercial crew?

In software, if you change one aspect of the code, everything has to be re-tested. Man-rated launch vehicles are just about that strict. Very minor changes might be permitted, such as adding a redundant pump to prevent a repeat of CRS-16, but vehicle-wide upgrades such as have happened over the history of Falcon 9 are verboten, for good reason.

Not only do you have countless parts on an orbital launch vehicles, each of them interacts with each other. For unmanned launches, if you do a thorough analysis of what a change might affect, you might get away with it. Manned launches are, for good reason, put to higher scrutiny.

Either SpaceX freezes the Starship design way earlier than they should, or they're not launching people on it for many, many years after its introduction.

[Rakaydos]

 I don't think you grasp the difference here. That hopper being built in Texas? It's replacement (with heat shield and brakerons) is supposed to come online just 3 months after it. In that time, the hopper is scheduled with the FAA to fly almost 40 times ... More than half as many flights as Falcon 9 (all versions) to date. Then the brakerons version will be testing at a similar rate. By 2020, there will be more Starship related test flights than falcon 9 flights.

As for NASA requirements, they don't matter if Yousaku and his artists are willing to sign a waiver with the FAA to go around the moon. Or if SpaceX employees are willing to sign the same waivier to build a glorified gas station on Mars. It matters what SpaceX can convince people of, not the government dotting every I, crossing every T and making sure the rest of the letters and numbers have serefs. 

[Starman4308]

Remember when the Space Shuttle was going to fly a gazillion times a year, and come in on time and under budget, and the US government was mandating that satellites be launched on the Shuttle? Remember when the Falcon Heavy was going to be ready to fly in 2011?

We've seen this dog and pony show before. Musk's timelines would be highly optimistic even without unforeseen events, like, say, a new booster design failing. Like they often do, something that would cause a significant interruption in schedule as the cause of failure is investigated, likely requiring redesign, causing the schedule to slip even further.

 

A 2023 manned launch mandates these things:

No new legislation to say "hey, we actually need safety standards now that NASA and Roscosmos are not the sole providers of manned launch services".

No unforeseen delays.

Starlink delivers on time, and customers actually buy their services.

No unforeseen snags in re-use of the Superheavy and Starship.

Nobody complains about the huge number of Starlink satellites, putting political pressure on Starlink/SpaceX.

Venture capital and loan funding comes in on time in sufficient amounts.

 

I put less than 1% odds on things working out this well for SpaceX.

EDIT: Whoops, forgot commercial crew delivery isn't there yet.

Edited January 22, 2019 by Starman4308

[tater]

I'm actually closer to @Rakaydos than not. I agree completely that crew won't likely fly until after many (Starlink) stats are delivered to orbit.

I'd not bet on 270 flights in that time period, but hopefully a large number. My argument was with it being airliner safe. It has taken airliners many decades to get to the current level of safety, and with many thousands of flights daily during all those decades, some of which found flaws which were corrected (either on existing aircraft, or improved in future designs).

I also actually think flying people is the killer app for rockets. It's the only way to get flight rates up that makes sense from a cost standpoint. I just don't see how they can demonstrate safety short of flying many, many flights.

Airline safety levels will need to be something people internalize, too, no amount of math will help if it's only flown a few hundred times, and people know the real rate is 1 in a million. They trust the latter number since it's maybe the number of flights that happen every week or so (the number of flights per year approaches 50 million), and crashes are rare.

Edited January 22, 2019 by tater

[kerbiloid]

Why launch the crowds while there is no space reactors on horizon?

A sixpack of humans in ISS uses those huge solar panels.
A hundred would require absolutely enormous solar panel area. No way to have them in orbit until a whole nuclear powerplant onboard.

Edited January 22, 2019 by kerbiloid

[tater]

43 minutes ago, kerbiloid said:

Why launch the crowds while there is no space reactors on horizon?

A sixpack of humans in ISS uses those huge solar panels.
A hundred would require absolutely enormous solar panel area. No way to have them in orbit until a whole nuclear powerplant onboard.

This is of course the fundamental problem of humans going anywhere in space. 100% of the solar system worth living in by humans is a built environment.

There is no real estate in the solar system unless someone builds it first.

I see the only crowds being sent for a long time as tourists, honestly. P2P, or even brief orbital trips, or excursions around the Moon and back would be worth it if the cost could ever come down to the level that even regular people who were just "pretty well off" could afford, not just billionaires. I'm thinking akin to First Class round trip airfare to the other side of the world prices (under 50k) for orbital, and maybe more like 100k+ for a lunar excursion.

If a lunar trip was reasonably safe, and 250k... I'd figure out a way to make that work before I die. For an 11 minute New Shepard or Spaceship One flight? I wouldn't be interested for 25k, I'd rent a house in Italy for a long vacation, instead, lol.

[kerbiloid]

20 minutes ago, tater said:

I see the only crowds being sent for a long time as tourists

And dealing with mass tourism we should face the fact that a proper LES would kill the random civil tourists with accelerations almost as effectively as a simple crash.

(If 20 of them die within a several years period after the failed launch due to the trauma, they will anyway be counted as victims)

So, looks like reliability and redundancy are the only way for multi-human ships. No LES, no capsules.

Edited January 22, 2019 by kerbiloid

[DDE]

7 hours ago, Starman4308 said:

Control glitch cuts the Superheavy thrust half a second into flight.

Not like that’s ever been a problem with 30-odd thrusters...

Edited January 22, 2019 by DDE
Replacing mislabeled photo of launchpad under construction

[Xd the great]

"In order to estimate the damage of a superheavy malfunctioning at a wrong time, 3800 tonnes of methane and LOX was detonated today. It was... spectacular"

[DDE]

42 minutes ago, Xd the great said:

"In order to estimate the damage of a superheavy malfunctioning at a wrong time, 3800 tonnes of methane and LOX was detonated today. It was... spectacular"

You have to practice Musk’s terraforming plans at some point.

3 hours ago, kerbiloid said:

And dealing with mass tourism we should face the fact that a proper LES would kill the random civil tourists with accelerations almost as effectively as a simple crash.

(If 20 of them die within a several years period after the failed launch due to the trauma, they will anyway be counted as victims)

So, looks like reliability and redundancy are the only way for multi-human ships. No LES, no capsules.

And that’s one of the reasons I doubt true, casual space tourism will ever blast off.

Edited January 22, 2019 by DDE

[Xd the great]

5 minutes ago, DDE said:

 

And that’s one of the reasons I doubt true, casual space tourism will ever blast off.

Suborbital ones will, just not orbital ones.

 

[Rakaydos]

8 hours ago, Starman4308 said:

Remember when the Space Shuttle was going to fly a gazillion times a year, and come in on time and under budget, and the US government was mandating that satellites be launched on the Shuttle? Remember when the Falcon Heavy was going to be ready to fly in 2011?

We've seen this dog and pony show before. Musk's timelines would be highly optimistic even without unforeseen events, like, say, a new booster design failing. Like they often do, something that would cause a significant interruption in schedule as the cause of failure is investigated, likely requiring redesign, causing the schedule to slip even further.

 

A 2023 manned launch mandates these things:

No new legislation to say "hey, we actually need safety standards now that NASA and Roscosmos are not the sole providers of manned launch services".

No unforeseen delays.

Starlink delivers on time, and customers actually buy their services.

No unforeseen snags in re-use of the Superheavy and Starship.

Nobody complains about the huge number of Starlink satellites, putting political pressure on Starlink/SpaceX.

Venture capital and loan funding comes in on time in sufficient amounts.

 

I put less than 1% odds on things working out this well for SpaceX.

EDIT: Whoops, forgot commercial crew delivery isn't there yet.

The difference is, SpaceX spent the last 8 years doing their homework. Remember the "I want to be able to launch with only 12 hours of touch labor" "we can cut it down to 24 on the falcon 9, but no further" argument? You can put money down that they were thinking about why that answer was true for the falcon when designing Starship/Superheavy.

If Superheavy has a teething failure launching Starlink, who's going to stop them? Oh, sure, they'll fix the problem for their own sake, but it's literally no one else's buisness what spaceX does with its own hardware. Design tools have improved, but SpaceX doesn't rely on simulations when they can rapidly prototype and test.

For the record, there are saftey standards. The FAA tracks them.

Starlink has a compelling buisness model, it requires a great many people to spite themselves for it to come online and still fail. Before SpaceX, launching was too expensive to try Starlink, but now that the buisness model closes, it will effectively print money, solving all of SpaceXs other problems.

 

[Xd the great]

6 minutes ago, Rakaydos said:

Starlink has a compelling buisness model, it requires a great many people to spite themselves for it to come online and still fail. Before SpaceX, launching was too expensive to try Starlink, but now that the buisness model closes, it will effectively print money, solving all of SpaceXs other problems.

 

What happens if starlink fails?

[Green Baron]

I'll have to rely on my slow satellite internet until eventually a cable end comes to me. Which may be after the end of the universe ..

[Rakaydos]

9 minutes ago, Green Baron said:

I'll have to rely on my slow satellite internet until eventually a cable end comes to me. Which may be after the end of the universe ..

Also Wall Street will be a few milliseconds slower trading with Europe and many miliseconds slower trading with Asia.

[DDE]

10 hours ago, Starman4308 said:

Nobody complains about the huge number of Starlink satellites, putting political pressure on Starlink/SpaceX.

You can cross that off the list. OneWeb alone, with its far less elaborate design, has made it onto Russia’s informal list of national security threats.

Edited January 22, 2019 by DDE

[tater]

3 hours ago, DDE said:

And that’s one of the reasons I doubt true, casual space tourism will ever blast off.

While I'm not sanguine about it, I think that tourism would be a boon to the space industry should they ever figure it out. This includes both P2P, as well as actual trips to space. This is for the simple reason that it's a virtually unending cargo.

7 hours ago, kerbiloid said:

And dealing with mass tourism we should face the fact that a proper LES would kill the random civil tourists with accelerations almost as effectively as a simple crash.

"Mass" is a relative thing. I'm not talking airline levels (unless they ever manage P2P), but far more than the current rate of 3 humans to space every 6 months.

 

 

It's interesting to consider abort modes on aircraft. The case of the Airbus that put down in the Hudson is a prime example of what planes can do with all engines out problems (that kills everyone in a Starship), but such forced landings are also pretty rare. Airline losses are really rare, but they tend to either be CFIG issues (controlled flight into ground---pilot error, in short), or something so catastrophic there is no second chance. Rockets eliminate pilot error, so all the systems then have to be made safe/redundant. I'm unsure how safe you can get rocket engines and their associated tanks and plumbing. Is there an upper limit?

[kerbiloid]

1 minute ago, tater said:

current rate of 3 humans

Specially trained volunteers, not random people from street.

[tater]

1 minute ago, kerbiloid said:

Specially trained volunteers, not random people from street.

Absolutely.

Yeah, I see early space tourism in larger numbers than the few super rich who have done so as being a sort of extreme sport. Safer than climbing Everest, but substantially less safe than flying to Paris to eat and go shopping. At least until they can demonstrate some operational safety. That's why you want to see unmanned test vehicles and cargo delivery. You want enough flights that they actually have problems, and ideally problems that get corrected on the fly and the craft still makes it. Engine out situations, etc.

Again, I see it as not easy, or even likely, but I do see it as a bottomless market should anyone ever figure it out.

 

[kerbiloid]

6 minutes ago, tater said:

the few super rich who have done so as being a sort of extreme sport.

Requiring a small capsule with a pilot, a secretary, a bodyguard, and a waiter.

Though, a multi-person ship full of billionaires is a thing, too...