Wayne Hale continues his remembrances of the tragic events of ten years ago. This is the nut of the piece:
Sometime after Bob left my office, Linda and I had another short phone conversation in which she told me that Bob was an excitable guy. I had to agree; he was pretty excited. But it seemed to be justified, rather than a reason to downplay the concern. Then she delivered the sentence that would define the rest of the tragedy; a sentence that was repeated as common wisdom by almost every senior manager that I talked to over the next two weeks: ‘You know, if there was any real damage done to the wing, there is nothing we can do about it.’ As unsettling as that was, I had to agree; going back to the first shuttle flight it had been well known that there was no way to repair the heat shield in flight. Nobody, not even me, thought about a rescue mission. Why would we?
It reminds me of what I wrote at the time (or rather, a few days after the Columbia was lost):
Imagine that you’re an engineer at JSC. The Shuttle is up, and there’s no way to bring it back except the way it normally comes back — a hot entry, just as it was designed for. There’s no other way of getting the crew out of it, and there’s no realistic way to get supplies to them to extend their mission to buy time until you can some up with some way to save them. If there’s a problem, you have no realistic options.
Now, you’re asked to make an assessment, in the absence of any data except a launch video showing some insulation hitting the vehicle, as to whether or not the damage could be catastrophic. Others around you, whom you respect, are saying that it won’t be. You have a bad feeling, but you can’t prove anything with the available data.
What do you do? What’s the benefit, given that there’s no action that can be taken to alleviate the problem, in fighting to get people to recognize that we may have a serious problem?
Moreover, suppose that we do believe that there’s a problem.
Do we tell the crew? What can they do, other than make peace with their God and say goodbye to their families? Think about the scene toward the end of the movie Apollo XIII.
“Gene, we think they may be entering a little hot.”
“Anything we can do about it?”
“No.”
“Then they don’t need to know, do they?”
It would make it difficult, if not impossible, for them to perform their experiments, knowing that they may be doomed at the end of it, and much of the results destroyed along with them, so if it turns out to be a false alarm, we ruined the mission.
It’s not hard for me to see how a group of smart people, all in the same situation, could reach a consensus that there’s not a problem.
The real problem is the fact that we send Shuttles off into the wilderness naked, with too few options.
This is also part of the theme of my book.
There wasn’t much anyone could’ve realistically done, but a few things do come to mind with the benefit of hindsight.
1. As soon as the impact was detected on the launch footage, request imagery of Columbia (either satellite to satellite or ground based telescopes) to include the underside of the vehicle at the earliest possible opportunity.
2. If a problem is detected, have them go to minimal energy consumption. Turn off everything not absolutely necessary for maintaining life support. See how long you can extend the mission duration.
3. Based on the results of step 2, consider what options, if any, are available to either rescue the crew or send them some more supplies to further extend the mission. Perhaps there was still nothing anyone could’ve done but making the best use of the resources on board and extending the time might’ve given them a better chance of surviving. Or not.
You can’t make good decisions without good information
Can it be definitively said that a resupply, after measures were taken to extend the mission duration to its maximum, was impossible?
I can’t imagine that we would not have moved heaven and Earth to do something if we had definitive information that the bird was fatally wounded.
What ever happened to “failure is not an option”?
I mean suppose, just suppose someone came up with an idea of a risky EVA and stuffing wet rags into the breach to form an ice plug? Maybe that wouldn’t work, but you could have let the crew decide what to do about their fate?
Maybe the Russians could launch a Soyuz with some kind of repair kit?
That last one wouldn’t be possible. Columnia was in an orbit with an inclination of about 40 degrees. The lowest inclination the Russians can launch a Soyuz capsule into is 51.5 degrees due to the high latitude of their Balkinour (Tyuratam) launch site and launch constraints. They don’t have enough delta-v in a Soyuz to lower the inclination enough to have rendezvoused with Columbia.
I think they might have been able to get to a lower inclination but only if China would be willing to absorb the second stage.
No, Baikonur is at 46 degrees inclination. Soyuz LV doesn’t have the yaw steering capability to get a Soyuz or Progress spacecraft to 39 degrees.
That should be “latitude”, not “inclination”.
The repair option was considered in the CAIB report, but the probability of success was considered to be lower than the rescue option.
The study was performed fairly quickly and did not have the benefit of much that was learned about TPS repair during the return-to-flight process, in particular the sensitivity of TPS heating to boundary layer transition caused by surface roughness. Based on that hindsight NASA would now rate the probability of success of the repair option even lower than it did at the time of the CAIB report.
The Atlantis rescue option was the only realistic one, and it was itself a longshot.
Perhaps more damning is the Challenger disaster. The launch should not have occurred on such a cold day — it was way below the limits that had been set for the shuttle.
Also of considerable interest is the fact that the engineers who made that decision were sleep deprived. Some of us know that really screws up thinking. Too many narrow engineers — and for that matter, narrow doctors — want to ignore the need for sleep.
We need more open minded people in engineering and other specialty fields.
John Logsdon, who was on the Columbia accident investigation board, reported the results to COMSTAC. What came as a surprise to me was the observation that Atlantis was in the VAB, ready to go, and could have been used to rescue Columbia had they known the extent of the damage.
That isn’t John’s unilateral opinion. It was part of the Columbia report.
Atlantis was in the OPF, not the VAB, and was by no means “ready to go” (see CAIB section 6.4, p. 173).
The CAIB report was not the last word on the subject; the section in question was based on a NASA study that was directed to use certain optimistic assumptions.
There was a Pegasus launched 9 days after the loss of Columbia, and a Delta launched 5 days after that, both from CC, but I suppose using either of those to resupply an orbiting and (in terms of orbital capabilities) fully functional Columbia would have been impossible for NASA.
Neither of those two were considered by the CAIB. The CAIB did consider an Ariane 4 that launched from Kourou on February 15 but offered no conclusions (CAIB D.13 section 5.2).
Keep in mind that none of the vehicles were capable of active rendezvous; Columbia would have needed to conserve enough consumables to power-up and perform the rendezvous. That would appear to rule out both the Delta and the Ariane, and the Pegasus would have been marginal.
STS-107 was not a rendezvous flight and did not carry rendezvous checklists, but checklists could have been uplinked. Husband and Chawla had rendezvous experience and McCool was rendezvous qualified.
In order to stretch the crew consumables that far, the crew would have discontinued exercise and would likely have been too deconditioned for a repair EVA. Therefore this option was less likely to succeed than the Atlantis rescue option.
They didn’t even try. This tells me NASA is the last organization that should be in the space business.
I vehemently disagree with the “nothing could be done so don’t tell them” scenario. That’s only arguable *IF* there was absolutely nothing the crew could have done to increase their chances for survival, even to a tiny degree.
In that light, it was worth finding out. If imaging the shuttle would have been viable, they ought to have done so IMHO.
For the sake of argument, let’s rule out launching a rescue shuttle, due to the issue of no one knowing if it, too, would be fatally damaged during launch.
That leaves Soyez, or self-rescue. I have no clue on Soyez (were two far enough along to be ready in time? You’d need two, launch them unmanned, and even so that’d mean 4 in one of them.).
That leaves self-rescue. Are we really saying that, even had the damage been known, there was nothing the crew could do to increase their chances even to a tiny degree? Alter the entry profile, or as a poster above says, stuff the hole with rags and add water to freeze it (making it more akin to pykrete than ice, so it would be harder, as well as take longer to melt).) Maybe something in the spacehab was made of titanium or stainless steel that could have been added to the top of the ice patch? Could we have been certain that doing something like that would not have increased the crew’s chances by even a tiny amount?
If there was anything that might have given them a better chance, however slight, not telling them would have been nothing short of reprehensible. I do see the point in the case of Apollo 13, where nothing could be done at that point, but not in this case.
The Columbia accident report outlines the ice patch idea as an option, as well as a rescue option.
http://www.pbs.org/wgbh/nova/space/space-shuttle-rescue.html
BTW, A Soyez launch out of Biakonur has it’s most efficient orbit at 45 degrees. They launch to 51.5 to avoid China.
Is there a significant payload loss be decreasing orbital inclination as opposed to increasing it during launch (relative the the launch site’s latitude)? Because if not, a Soyez could launch to 39 degrees (Columbia’s inclination) as easily as it launches to an ISS inclination of 51.5.
If they had a pair of scissors maybe they could’ve cut some of the fibrious insulating blankets and nomex insulation from much less critical areas and used it as a hole stuffer, perhaps buying some time and even ablating a bit.
The lowest inclination you can directly launch into without some complex maneuvering is the same as the launch site latitude. You get that by launching due east. Any other launch azimuth will cause an increased inclination. I’ve seen some launch profiles where the booster flies a dogleg during ascent at the expense of some payload capacity. However, I’m pretty sure the Soyuz booster doesn’t sufficient surplus capacity to launch a Soyuz capsule into a substancially lower inclination.
As for your “don’t tell them” argument, I fully agree. Back in the Apollo moon landing days, there was concern that the Lunar Module ascent engine wouldn’t ignite, stranding the astronauts on the lunar surface until they ran out of oxygen. When asked what he’d do if that happened, one astronaut replied that he’d work up till his last breath trying to solve the problem*. The Columbia astronauts would’ve done the same.
*I’ve read that when Apollo 11 was getting ready to lift off from the lunar surface, a key toggle switch needed for the launch sequence broke. From what I’ve read, Buzz Aldrin jammed an ink pen into the switch to make the contact. You can be pretty inventive when you’re ass is on the line.
Is there a significant payload loss be decreasing orbital inclination as opposed to increasing it during launch (relative the the launch site’s latitude)? Because if not, a Soyez [Soyuz] could launch to 39 degrees (Columbia’s inclination) as easily as it launches to an ISS inclination of 51.5.
Nice try, but a 51.5 degree orbit does pass over locations at 45 degrees North, a 39 degree orbit does not passes over locations at 45 degrees. it’s expensive of delta V to get to an orbit that doesn’t pass over your launch site.
“Nice try, but a 51.5 degree orbit does pass over locations at 45 degrees North, a 39 degree orbit does not passes over locations at 45 degrees. it’s expensive of delta V to get to an orbit that doesn’t pass over your launch site.”
Ah!!! Thanks!
That means a Soyez could not have reached Columbia’s orbit (nor could Columbia have reached 45 degrees).
Like Dennis Wingo said above if it was determined they had catastrophic damage I am sure we would do what ever it takes. Given unlimited resources someone somewhere may have been able to formulate a rescue plan. Unfortunately we will never know.
The CAIB report addressed the possibility of crew rescue. Was it possible? Yes. Would it have happened? Doubtful. But those discussions miss some basic problems caused by the mindset that began: “if there was any real damage done to the wing, there is nothing we can do about it.” That’s a lie, and one that needs to be entirely put to bed.
First of all, imagery is far from nothing. Obtaining the imagery was and is incredibly easy to do. That the request for additional imagery was denied is horrific and should have cost Linda Ham her job. Would imagery have saved the crew? No. But her decision to deny the imagery request kept any further action from being accomplished, and cost NASA (the US taxpayers) millions in just trying to determine by analysis what happened.
Second, and this is the larger problem with “there is nothing we can do about it.”; they brought the wounded vehicle over some of the most populated parts of the United States. It is stunning to consider the footprint of debris left by Columbia and that no one on the ground died. The fact nobody on the ground died has nothing to do with the leadership exhibited by NASA during STS-107. Again, had imagery been taken, its likely it would have lead to a conclusion that the damage was fatal for the orbiter. With such information, does anyone believe we would have brought in Columbia on that orbit or to KSC? That’s a decision that could have been made, with or without imagery, if you simply had a belief that something more could have been done.
Heck, all you needed is interest in something a bit more broad than the orbiter itself. NASA’s risk matrix is a 5×5 with top severity as loss of vessel/loss of crew. Once they figured they hit the worse severity, that’s why NASA leadership stopped thinking anything could be done. In other industries, the top of the matrix isn’t just loss of life, but loss of life or effects beyond the fence line (ie your disaster affects people near you, not just you). That simple change in mindset might have pushed NASA leadership into considering the safety of the people Columbia flew over during entry. Yet, by the end of the Shuttle program, that aspect of risk wasn’t changed. That’s simply irresponsible yet entirely expected. You don’t respond when you believe there is nothing you can do about it.
Leland, the issue with the re-entry track and danger to people on the ground as a reason to find out if there was damage is something I’ve never seen raised before. I think it is a most excellent point, especially as the breakup was seen over Dallas. That means Dallas was on the ground track, so had the shuttle broken up just a little earlier, Dallas would have been hit, and hit hard. They were very lucky that most of the footprint was in sparsely populated areas, and even so, there were several very close calls.
They may well have still chosen KSC if they thought a breakup was probable but not certain, though I’m sure the track would have come in over the gulf, not Texas, and sure as heck not over a major city. Edwards would be more likely IMHO, on a track that avoided heavily populated areas. Or more likely still, an abort site with an all-sea track. (Vandenburg?) They’d have a further problem in that they wouldn’t have wanted to try using any cross-range ability with suspected damage.
My personal guess is that, had they known of the damage early in the flight, they’d have gone to long-duration procedures while rushing Atlantis for launch. I think this decision would have been made for political rather than logical reasons; I can’t see a president (and he’d end up being the one to make the decision) let seven astronauts die in what would be a media spectacular. Logically, risking another orbiter and 4 more crew would not be a wise call IMHO (when you had no idea if it, too, would arrive in orbit fatally wounded) but I’m betting that’s what they’d have done. They’d have rolled the dice. I don’t see not doing so as being a politically viable option, and politics trumps common sense every time.
And in this case, they’d have probably succeeded, because as we now know (and didn’t until well after Columbia) it was statistically unlikely that a bipod ramp would come off and then be slammed into the wing leading edge on any given flight. (had it come off much earlier or later in the ascent, it probably wouldn’t have done much damage).
CJ,
For me, the issue was very clear. My wife and a daughter were visiting my father-in-law in East Texas that morning. They all heard the sonic boom as debris started coming down. They were at the west end of the debris field. On the east end of the debris field was the home my parents just purchased. Those are loved one put in harms way by the poor decision.
As you say, I don’t know if they would have avoided KSC, but they would have come in from the ascending orbit rather than descending. I’d think Edwards as well for the dry lake bed, but in early February, I’m not sure that’s an option, and moreover, you threaten LA. So then an abort location would be the next call.
Also, I’ve always wondered how much further Columbia would have travelled if Husband delayed turning off the RCS jets that were compensating. That might have got them over the Gulf, or if the propellant held out, maybe over the Atlantic. But I’ve been told that the temp sensors deep in the wing were already spiking and dropping off, so it may not have bought much time.
I do think long duration may have occurred. I just don’t know, because the mindset just wasn’t present. I just don’t know how to convey this to people like NEMO. The next step after declaring long duration is preparing Atlantis, which includes convincing everyone that the next ET is some how safer. When you think of “there’s nothing more we can do”, then how in that short time do you get people to think about how to make the next ET flight safer. And if anyone thinks we licked the problem of ET foam shed (or ice) they missed STS-125, STS-127, and STS-134 (although all had a common problem with their ETs, which we knew before STS-134). With that understanding, I’m uncertain as to whether NASA would have launched the rescue. Then again, the notion that they would have gotten the data would demonstrate a different mindset.
BTW, I agree statistically, it would be worth the risk. We still had foam shed, but the odds were still near 1/100.
As far as how the crew would react. Well, we already have a clue: “I finally gave in to Hoot’s solution. The day before (entry), as he floated to the windows to do some sightseeing, he said, ‘no reason to die all tensed up.’ I would do my best to relax and enjoy the sights.”
STS-27R had an arm, STS-107 did not. However, I’m 100% confident that we could have gotten other imagery, if only management had simply asked. There’s a rebuff made in the CAIB report, but what is written is far more mild than the reality. The decision not to get imagery was made with complete ignorance. What I know of it is really horrifying.
False. The post-Columbia flight rules required entry overflight risk to be minimized if the orbiter TPS was known to be compromised.
That’s a flight rule limited to a singular threat. I’m sorry you didn’t comprehend what I wrote. I’ll try and restate it again.
The top level severity for risk for SSP was loss of vehicle/loss of crew. It did not consider damage to things beyond NASA. That mindset drove the notion that if the damage resulted in loss of crew, nothing more could be done.
Flight rules are good for known threats, but few are written for unknown threats. In fact, many flight rules were written based on lessons learned from failures with the fortunate aspect of many of those failures occurring during sims. What happens when a new failure previously unseen occurs? For NASA, they go through the risk assessment process I described. That process was limited, and that limitation drove some of the poor decision making during STS-107. It’s why the flight rule was written after STS-107 and not before or during.
OK, my previous assessment was that your statement was wrong. My current assessment is that your statement is utterly meaningless in any real operational sense.
That’s fine. One of us actually had a role of determining whether the TPS was compromised, signed off on the flight rule you mentioned, and knows the difference between the flight rule process and risk assessment process at NASA and on the Space Shuttle Program. I’m sorry I can’t help you understand this difference, but I do recognize that for many its tough to see the distinction. I think it is probably one of Rand’s motivations for writing his book.
A Flight rule is an instruction you take to prevent a risk. Such instructions are developed because of a hazard or risk analysis done previously. New anomalies happened all the time during the program, and new flight rules were created later. Between the anomaly and the flight rule, a hazard analysis was performed. When a hazard analysis comes up with “nothing more can be done”, then sometimes that prevented good flight rules from being written.
It’s mindset. A tank of foam for repairs would be nothing to include. Do they have fire extinguishers?