In reply to j_tso :
Might want to look a little further back in history. NASA popped a few in the early days.
In reply to j_tso :
Might want to look a little further back in history. NASA popped a few in the early days.
In reply to Keith Tanner :
I mean NASA now, when we we're not trying to beat Russia to the moon.
In reply to j_tso :
You mean, when they were learning how to actually make these new technologies work. The race to the moon was just the pressure to get it done.
SpaceX can perform tests that may end in what looks like a failure, because they're engineering tests as they push boundaries. They're not putting people or valuable cargo on board, and they don't have a bunch of politicians getting nervous about "wasting" money.
Their hardware rich, rapid iteration methodology isn't what we're using to seeing today from NASA (which has blown up a few ships WITH people on them for bad reasons since the moon race was over) but it gave us the Falcon 9. The F9, in case you're not aware, is one of if not the most reliable rockets ever. The current version has had 162 launches, every one of which was a success. It's also the only reuseable orbital booster and has had well over 100 successful landings. It's also the rocket NASA relies on to get people and a large portion of supplies to the ISS. So maybe they don't blow up all that much stuff after all...
I'll bet a lot of people watching this Starship pop weren't aware that there was an F9 launch yesterday and there are going to be 4 more before the end of the month. They just work, they're not newsworthy anymore.
I'm reminded of when the Abrams tank was under development, with its turbine engine, and other new-to-tanks systems.
They would run grueling tests on the prototypes, and they would break, or operate differently than expected, or whatever.
And every time, the media would shout EXPENSIVE NEW TANK A FAILURE.
I remember one of the brass on the project basically saying, "You numbskulls, these aren't failures. Finding out what breaks is the whole point of testing so we can make those parts better."
I noticed that they had 3 motors not functioning at ignition, then eventually lost up to 6 for a while before finishing with 5 shut down. I wonder what kind of escape system they have on the Starship crewed second stage? I hope that the crew doesn't just have to ride it out like the Space Shuttle where if the crew cabin had been a safety capsule imbedded into the ship and been equipped with a parachute, at least one of the lost missions could have been saved.
I think that Space X has done a phenomenal job with their reusable launch systems. I just can't take Ellen, and I call him that because that is how I thought you would pronounce his name for the last 15 years until I finally heard someone say it on TV recently.
From what I've read, there are two schools of thought on how to do rocket development. The one that most people use these days is very slow and methodical, making small changes to stuff that's already known to work. It takes a long time and costs a lot of money to really improve on the current state-of-the-art, but it's "safe" in the sense that it minimizes the chances of big splashy failures in the news. SLS is the epitome of this -- they took existing shuttle designs, moved the engines to the bottom of the external tank, and strapped SRBs with an extra section in them to the side. That took 11 years (15 if you count the very similar Ares V development that happened before it), cost fifty billion dollars, and has gotten us 1 launch so far (albeit a successful one).
The other school of thought is to be aggressive, to make riskier launches more often. This is what SpaceX does, and it gets stuff done faster and ultimately at lower cost but has PR costs in terms of public failures. Starship was started in 2016 or, is a completely new (and radically different) design, and is estimated to have cost between 2 and 10 billion (figures aren't public because they're a private company). They haven't had a successful launch yet, but even if it takes another 4 years and 40 billion dollars it will still be faster and cheaper than SLS.
SpaceX had 3 failures of their first rocket, Falcon 1 (the first three flights), and 2 of Falcon 9 (one kaboom in flight, one on the pad during fueling, both pretty early in the program). The explosion footage roll above is mostly landing attempts, which was R&D being done after the successful completion of the launch. That's a pretty good way to do R&D -- you get the customer to pay for the expensive part of all of your test launches (and happily so because their payload makes it to orbit) and then you do the landing attempt with a rocket that was just going to get dumped into the ocean and thrown away otherwise.
Keith Tanner said:SpaceX can perform tests that may end in what looks like a failure, because they're engineering tests as they push boundaries. They're not putting people or valuable cargo on board, and they don't have a bunch of politicians getting nervous about "wasting" money.
That's the point I failed to add in my first response. Today's NASA isn't allowed to have hiccups because there's a contingent in the government that wants to cut everything. If the SLS was not only delayed but also exploded they'd be restricted to weather balloons.
In reply to Duke :
Not to mention the F-22 Raptor that congress cut funding for after only 195 out of a proposed 750 where built leaving the US without enough air superiority fighters to protect our interests.
VolvoHeretic said:I noticed that they had 3 motors not functioning at ignition, then eventually lost up to 6 for a while before finishing with 5 shut down. I wonder what kind of escape system they have on the Starship crewed second stage? I hope that the crew doesn't just have to ride it out like the Space Shuttle where if the crew cabin had been a safety capsule imbedded into the ship and been equipped with a parachute, at least one of the lost missions could have been saved.
The Challenger crew might have been able to survive with an escape system like that, but it's unclear if it would have been possible to build one into a Shuttle, it's a much more complex requirement than the systems used on F-111s and the like and space hardware is even more weight-sensitive than aircraft.
My impression is that it's going to be a long time before they put people on a Starship. I think to start with it will be a cargo vessel, with crews going up on Falcon 9/Dragon and rendezvousing in orbit.
One of the advantages to having multiple engines on the rocket is that individual engine failures are less significant because there's a lot of redundancy. It does significantly increase the complexity and require a sophisticated control system to manage the failures, though. The Soviet "N1" moon rocket used 30 engines in the first stage, they tried four launches in the late 60s/early 70s, all of which failed due to these sorts of complexity problems.
I am curious as to how a Starship abort would work. With Dragon, they've proven (with an actual flight) that the capsule can pull itself away from an exploding booster. But that's using hypergolics, I don't think the Starship engines could spin up in time to make that a possible option.
Keith Tanner said:I am curious as to how a Starship abort would work. With Dragon, they've proven (with an actual flight) that the capsule can pull itself away from an exploding booster. But that's using hypergolics, I don't think the Starship engines could spin up in time to make that a possible option.
Also, the Starship engines are optimized for vacuum and it likely has a much lower thrust to weight ratio than Dragon does on the dracos.
One big ol' takeaway on that launch was that flame trenches and water deluge suppression setups are a *good* idea to have as part of your launch pad... https://gizmodo.com/spacex-starship-launch-pad-damage-video-1850357836
codrus (Forum Supporter) said:VolvoHeretic said:I noticed that they had 3 motors not functioning at ignition, then eventually lost up to 6 for a while before finishing with 5 shut down. I wonder what kind of escape system they have on the Starship crewed second stage? I hope that the crew doesn't just have to ride it out like the Space Shuttle where if the crew cabin had been a safety capsule imbedded into the ship and been equipped with a parachute, at least one of the lost missions could have been saved.
The Challenger crew might have been able to survive with an escape system like that, but it's unclear if it would have been possible to build one into a Shuttle, it's a much more complex requirement than the systems used on F-111s and the like and space hardware is even more weight-sensitive than aircraft.
The Challenger crew survived until the impact with the ocean. If only they had had a ballistic parachute.
Bing.com: Search: challenger disaster crew alive until impact with ocean
VolvoHeretic said:codrus (Forum Supporter) said:The Challenger crew might have been able to survive with an escape system like that, but it's unclear if it would have been possible to build one into a Shuttle, it's a much more complex requirement than the systems used on F-111s and the like and space hardware is even more weight-sensitive than aircraft.
The Challenger crew survived until the impact with the ocean. If only they had had a ballistic parachute.
Bing.com: Search: challenger disaster crew alive until impact with ocean
The fact that they survived to hit the ocean was up to chance, and depended on the specifics of how the shuttle broke up. To do an escape capsule right you need to take random chance out of the equation as much as possible -- the capsule needs to separate from the rest of the vehicle in a predictable manner, keeping all of the bits it needs and leaving behind all of the bits it doesn't. You need a system to open the parachute at the right time (avoid melting it with the rocket exhaust, fall quickly down to an altitude where there's breathable air, but obviously high enough to slow down enough), and you need a flotation system so that it doesn't immediately sink to the bottom of the ocean with the unconscious/injured crew still in it. You need beacons so that the rescue crews can find the capsule easily, a means of opening it easily so that the rescue crews can get them out without flooding the capsule, etc, etc. Even then it would have been useless for Columbia unless they'd put another heat shield on it with enough passive aerodynamics to ensure it was pointing the right direction. That's all doable if you design it in from the start, but retrofitting it to the shuttle was basically impossible -- you're better off starting over with a new vehicle.
And by the time you've done all of that, your escape capsule looks a whole lot like an Apollo-style command module, and that's why Dragon is a return to that style of vehicle. From a safety standpoint the Shuttle's design was flawed from the start.
As for how you solve that with Starship, one option as I mentioned above is to launch it empty and rendezvous in orbit. Crew goes up in Dragons with their abort modes, transfers to Starship in LEO, and goes off to the Moon or Mars or wherever in the Starship, leaving the Dragons in earth orbit. Maybe you rendezvous again to land at the end of the mission or something.
In reply to codrus (Forum Supporter) :
It's only cheaper if it works. It's more expensive when it doesn't. Which is why the methodical approach is a lot more common.
j_tso said:Keith Tanner said:SpaceX can perform tests that may end in what looks like a failure, because they're engineering tests as they push boundaries. They're not putting people or valuable cargo on board, and they don't have a bunch of politicians getting nervous about "wasting" money.
That's the point I failed to add in my first response. Today's NASA isn't allowed to have hiccups because there's a contingent in the government that wants to cut everything. If the SLS was not only delayed but also exploded they'd be restricted to weather balloons.
The problem is that all of these failed costs get passed onto us, the ones who eventually pays for all of this adventure.
Keith Tanner said:I'll bet a lot of people watching this Starship pop weren't aware that there was an F9 launch yesterday and there are going to be 4 more before the end of the month. They just work, they're not newsworthy anymore.
To me, that's exactly why it is a big deal, more so that they try to pass it off as a total success.
alfadriver said:It's only cheaper if it works. It's more expensive when it doesn't. Which is why the methodical approach is a lot more common.
Well, as I mentioned, they've spent less than 10% of what SLS has by this point. It'll take a lot more than one test launch failure to wind up being more expensive.
Keep in mind that SpaceX is building Starships and Super Heavy boosters faster than they can launch them as they fine tune the factory. It's either launch or scrap. That's exactly why they launched a few of the 10km flights. There was a known problem that made the chance of a good landing unlikely, but there's a whole bunch you can learn in the rest of the flight. Of course, all people remember is the failed landings. The first next-Gen ship that launched nailed it.
Is it more expensive to test often and allow for failure? It appears not. Is it less common? Not in what's referred to as New Space, the group of companies that are bringing new rockets and tech to market.
The slow and methodical approach is taken by established companies that aren't doing anything new, and they are notoriously more expensive because cost is simply not a factor at all. Look at how much more Boeing got paid for Commercial Crew, and the bids for the HLS contract.
The only reason anyone might call todays test a TOTAL success is because they can only comprehend black or white. It wasn't. But you also can't call it a total failure. It was more successful than not, because getting that beast off the pad was the biggest question.
If you're worried about the costs being passed in to customers, you should be extremely happy with SpaceX's approach. It's what got us the F9, which has brought the cost of putting 1 kg of material into orbit from over $50k (shuttle) to under $3k.
Keith Tanner said:...which has brought the cost of putting 1 kg of material into orbit from over $50k (shuttle) to under $3k.
$3000? That's cheaper than a funeral. I might have to try that when I check out.
I will say that Elon's "go fast and break E36 M3" ethos seems to work better with SpaceX than it does with say, Telsa's autonomous tech. The main issue being that it's being "beta tested" by random people with no training on public streets with innocent bystanders. In fact, overall I like SpaceX Elon much better than either Tesla Elon or Twitter Elon.... 
Tom_Spangler (Forum Supporter) said:I will say that Elon's "go fast and break E36 M3" ethos seems to work better with SpaceX than it does with say, Telsa's autonomous tech.
I'm not an aerospace engineer, but my impression is that's at least partly because SpaceX's challenges are more in the realm of "nobody has tried this before" rather than the "we've been thinking about this for 60 years and STILL don't have a clue how it would work" that exists in real AI.
If you purposefully test something to failure, it is not failure. Whether you are testing age, performance, or whatever, until you go over the cliff, you do not know where the cliff really is. If you don't know where it really is, you don't know how close you are too it. Not testing to failure leads to more hedged bets, which leads to higher operating costs.
The chain is as strong as the weakest link. There are too many variables to find the weakest through analysis alone. Until you find it, every other link is carrying extra weight. At the height of shuttle, wasn't the cost to orbit something like $8000 per pound? Extra weight is a cost passed to the customer.
Tom_Spangler (Forum Supporter) said:I will say that Elon's "go fast and break E36 M3" ethos seems to work better with SpaceX than it does with say, Telsa's autonomous tech. The main issue being that it's being "beta tested" by random people with no training on public streets with innocent bystanders. In fact, overall I like SpaceX Elon much better than either Tesla Elon or Twitter Elon....
Probably because Elon is so busy making an ass of himself wrestling with Twitter that he doesn't have much opportunity to interfere with SpaceX. So the people there who actually know what they're doing can actually do their jobs.
You'll need to log in to post.
