Did they make any estimates on the chance of success vers. failure on the moon landings? How dangerous did they consider the first Apollo missions at the time? And how dangerous are they estimated to have been, with today’s knowledge? Is today’s estimate considerable different from the estimate in 1969? Would the odds be acceptable today? Were the astronauts foolhardy, or had they the odds on their side?
IIRC they did, and with the Shuttle as well. If you take all of the malfunctioning and/or deadly space accidents all together (US and Russia), you get something like a 3% failure rate.
3 astronauts died during the Apollo programme. Grissom, Chaffey and White died in Apollo 1. There were 12 manned Apollo missions, each with 3 crew. So out of 36 crew, 3 died, which is a ~9% mortality rate (by my reckoning). How does that compare with the shuttle?
Apollo was more dangerous than those stats would suggest. You also have to consider the close calls - Apollo 13 came within a whisker of losing the crew. On Apollo 11 the guidance computer for the LEM failed. Other non-moon Apollo missions had various failures that could have been life threatening.
Apollo was a higher-risk program than was Shuttle. There was less engineering data to work from, it was a crash program, and everyone just accepted the risks.
In terms of actual risk assessment, the Shuttle Probability Risk Assessment Number (PRAN) is about 1/100 (that the probability of complete loss of crew and vehicle for any given mission). But that number needs to be taken with a grain of salt - before Challenger, NASA thought it was about 1/4000.
For Apollo, when the program started the estimates of a launch, landing on the moon, and safe return were pegged at about 2% (that’s a 2% chance of success, not failure). NASA worked hard to improve that number throughout the program, but it was never very high - nowhere near shuttle safety estimates as far as I know.
I have no data to supply, but having read more than my share of Apollo memoirs, everyone in the program assumed astronauts would die before we reached the moon, although the way in which Grissom, White and Chaffee actually met their end was largely unanticipated. (Obviously – if it had been, they wouldn’t have run those tests in so hazardous a manner.)
–Cliffy
The problem with assessing the risk of Apollo is that our data set is so limited. There were just eleven actual space missions: Apollos 7 through 17. Apollo 7 didn’t even leave Earth’s orbit, and 8-10 went to the Moon but didn’t land. 13, of course, intended to land but didn’t get the chance.
An assessment of Apollo as being relatively safe (I know three men died in “Apollo 1” but it was a stupid ground accident, not a spaceflight accident) has to be based on the assumption that 11 space missions is an accurate data sample. That’s not a reasonable conclusion. 11 missions ISN’T much of a data sample. If a mediocre quarterback happens to complete 11 passes in a row you don’t assume he’s suddenly become Dan Marino, you assume he got lucky.
It’s quite logical to think that they may have just gotten lucky. I think, personally, that if you could rewind time and do the Apollo program over again there’s a good chance one or more missions would end in catastrophe - as Sam points out, they lucked out big time with Apollo 13 and had other serious glitches as well. But, thankfully, they hit an inside straight. Sometimes you get lucky.
I saw a talk by Burt Rutan, where he gave statistics on the fatality rate of early flight. It was actually higher than that of the space program. Rutan said that one reason we are so slow about advancing into space is that we are so cautious, which might be because it is public, not private.
Rutan’s right. Or put in the cruder terms he’s used before - “The problem with NASA is that it’s not killing enough astronauts”.
Experimental test flight is, and should be, dangerous. You want to see a high fatality rate? Go look at how many test pilots died during the dawn of the jet age.
Or look at other risky aviation ventures. For example, the Blue Angels demonstration team has lost 10% of its pilots - 26 out of 260 - to accidents either in training or during airshows. The Reno Air races have had 18 fatalities. If we can tolerate a 10% fatality rate for airshow performers, we should be able to tolerate at least as much for people pushing the far reaches of the envelope for activities which benefit all of humanity.
The alternative is to make our research programs far more complicated, expensive, and slow.
I have a minor quibble with this statement, stemming mostly from how we would define a “whisker”. Then I’ll address the OP’s question, which is a fascinating one to me.
I’m going from memory here, mostly from Lovell’s book “Lost Moon”, and Andrew Chaikin’s “A Man on the Moon” (sorry no firm cites, I’m in the middle of a big project).
I recall Apollo 13’s situation being compared to a sky diver who pulls his rip cord and the chute doesn’t deploy. He then pulls the backup chute and it does deploy.
Not a good situation, to be sure. But that person is operating on a known backup system, with good chances of a safe outcome. If something ELSE were to go wrong, then it’s a fatality.
Most of the system failures on Apollo 13 had been planned for. For example, despite the scene in the movie, the improvised carbon dioxide scrubber was not invented on the spot. That was an established procedure, albeit one in the back of the book that nobody thought they’d ever have to use.
Same for using the Earth as an aiming point during course correction. I think Lovell had even participated in creating that procedure. Indeed, the whole concept of LM as lifeboat had been around since the Mercury days.
This is not to say it wasn’t a bad situation for 13. But the real issue was that all of these failure modes happened at once, in a combination they hadn’t envisioned. It had been assumed that a catostrophic failure of more than one system would kill the crew. In this case it didn’t. So they were left to implement backup procedures that were known, but under time pressure.
I remember somebody in Mission Control saying that after the situation had been stabilized somewhat after the initial onslaught of problems, they didn’t have any real doubts about bringing them back in safely.
So yeah, it wasn’t good. I wouldn’t have wanted to be there (and I’m a professional pilot who loves flying adventures). But not sure if I’d call it by a whisker either.
As to the general question of how dangerous Apollo was, sure it was dangerous. But it was carefully planned out for the most part, despite some glaring errors.
The whole program was done quickly. Had it not been on such a tight time schedule we might have done it much differently. And it was enormously complex. Landing on the moon and returning depended on a huge, complex chain of events to happen in sequence and on time. Frankly, I’m amazed it worked as well as it did.
Despite preferring not to be on Apollo 13, I’d volunteer to go in a heartbeat even if the odds were twice as bad as history leads us to believe.
NASA isn’t a single monolithic whole. It’s my understanding that before Challenger, the NASA administration talked about a 1/4000 rate, but most of the actual engineers did say something more like 1%.
Hmm. But, to be totally heartless and leave the human factor out, the space program costs SO MUCH more than any of the other activities mentioned. Estimates, in modern dollars, are that a single Saturn V launch cost $2.4 billion. Just losing ONE vehicle is a financial cataclysm in comparison to any other flight loss. I doubt that decreasing astronaut safety would cut that cost significantly; rather, the cost of astronaut safety could be looked at as insurance: that astronauts would be around to do their part to ensure that the expensive vehicle would not be lost. Any safety measure along the way increases the mission success rate, and protects your investment.
You’re forgetting 9 was an Earth-only too. First check out flight for the Lunar module.
The scariest parts of the missions were probably the multiple engine starts and restarts. Restarting may be more reliable today but back in the late 60s rocket engines either failed to start or blew up on start with disturbing frequency. Consider the overall mission profile and the consequences of engine failure/explosion at each step:
- Launch. Launch escape possible early in flight. Possible abort-to-orbit or abort-to-reentry modes later on. Not something you’d want to have to try.
- Trajectory insertion: engine failure means return to Earth. Engine catastrophe means probable loss of crew.
- Lunar descent. Failure of LEM lower stage to start means return to Earth. Premature shutdown means loss of lunar crew. Maneuver failure means loss of lunar crew.
- Lunar ascent: Failure of LEM upper stage means loss of lunar crew.
- Return trajectory: Failure of CSM engine to restart means loss of crew.
- Reentry: Any guidance or maneuver failure means loss of crew.
My overall impression is that the Apollo lunar missions were comparable to three men crossing the Atlantic in an open rowboat: well-trained well-prepared people can do it and have done it; but it’s VERY risky.
But the lucky part was that the damage was constrained to things they could fix. Apollo 13 exploded in space. The explosion was severe enough to blow the panels off the side of the craft and ruin the main engine nozzle. In a craft as lightly built and tightly packed as that, it seems remarkable that so much damage could be done and yet be isolated to systems that they were able to work around.
Absolutely. They really built those things, didn’t they?
I just realized an interesting irony. This thread got me surfing the web (in spite of my big project) and I came across Penn & Teller’s Bullshit episode about NASA.
They’re two hardcore libertarians who openly hate government. But they’re also big space enthusiasts. So they grudgingly conceded that NASA, despite its problems, has accomplished truly amazing things. Especially for a government operation.
Of course, they went into detail about NASA’s problems and failures. But they did not dispute the fact NASA is an example of government that does get something done.
Thousands of separate systems,millions of parts all made by the lowest bid contractor trying to cut costs as much as possible to make money. What could go wrong?
The LEM and CSM used hypergolic fuels, and therefore a start was virtually
guaranteed unless the valves didn’t open. Were there other reliability problems with them? Valve jamming? ‘Hard Start’ problems?
I once read that solar flares were a big concern. If there had been flares during an Apollo mission, the astronauts would have received a lethal dose of radiation. There was no practical way to shield the spacecraft, so the plan was “hope we don’t have any solar flares for the next twelve days.”
I have no knowledge of the designs so I couldn’t say; I was speaking generically of the consequences of engine failure for any reason. I have heard that the LEM engines especially were designed to be “too simple to fail”. I’m also put in mind of a science fiction story I read too long ago to remember details, but one quote stuck with me: “A ship that could not move was invariably a doomed ship”.
Maybe you’re too young to remember when launch vehicles blew up (or had to be blown up) more often than not. They were pretty expensive too. When you do something new, you have to expect to loose launch vehicles and crews.
On the other hand, politically NASA was probably right.
All I know is that failure was not an option.