NASA releases Columbia report

[Suzy]

Columbia Crew Survival Investigation Report (16.2 MB PDF) - 400-page NASA report of more details of the breakup and what happened to the crew.

1.1.1 Events with lethal potential
There were five events identified with lethal potential to the crew.

The first event with lethal potential was depressurization of the crew module, which started at or shortly after orbiter breakup.

The second event with lethal potential was unconscious or deceased crew members exposed to a dynamic rotating load environment with nonconformal helmets and a lack of upper body restraint.

The third event with lethal potential was separation from the crew module and the seats with associated forces, material interactions, and thermal consequences. This event is the least understood due to limitations in current knowledge of mechanisms at this Mach number and altitude. Seat restraints played a role in the lethality of this event.

The fourth event with lethal potential was exposure to near vacuum, aerodynamic accelerations, and cold temperatures.

The final event with lethal potential was ground impact.
​

 

[Thunder Chicken]

This was an interesting report, particularly for the accident timeline. It is good that they are doing this, maybe they can design spacecraft and procedures to increase survivability when things go wrong instead of hoping that they won't go wrong in the first place.

The parachute automatic release issue was interesting (as I used to pack parachutes). They can't use a conventional automatic release as they typically are set for density altitude (as you don't want your parachute opening at 100,000 ft, which would mean your bottled O2 would be gone before you reached a breathable altitude). Apparently the only way to arm the releases was to use the bailout boom. I guess it doesn't make much difference.

Reading the report, I began wondering if some sort of in-seat MOOSE setup might have helped, as it seems the tumbling accelerations were survivable. Assuming you are in a situation that vehicle breakup is going to occur, automatically close a tough Kevlar/foam clam shell around each seat that can allow the crew member to survive the breakup and thermal loads, and either deploy a canopy or kick the crew member out and automatically deploy their parachute at 14,000 ft MSL. It assumes complete breakup of vehicle, but it would be much lighter than an ejection system and could be used on the middeck, and it would increase your chances, maybe not to 100%, but better.

It sounds scary, but maybe the thing to do would be to design the spacecraft to break apart in a certain manner (something like an F1 car) when the worst happens.

 

[Andy44]

The seat idea I had also thought of, but I imagine that would be prohibitive in terms of weight and space taken up. Besides, there's no guarantee that in any given accident, you will find yourself flung free of the cabin. Challenger's cabin stayed intact until surface impact.

Perhaps a better idea would be to make the crew ride launch and re-entry inside a portion of the cabin which is also a detachable capsule. Not a new idea, of course, and also a big weight penalty. Plus, it defeats the original purpose of making a space shuttle that is comfortable and easy to fly and almost as safe as an airplane.

I've been thinking of starting a "what-if" thread in which readers are asked to design a space shuttle with similar capabilities to STS, but incorporating lessons learned.

 

[RocketMan_Len]

I've been thinking of starting a "what-if" thread in which readers are asked to design a space shuttle with similar capabilities to STS, but incorporating lessons learned.

Go for it! I've been working on a few ideas of my own...

- a two-stage VTHL ...
- first stage is a flyback vehicle
- both stages use non-cryogenic propellants
- orbiter equipped with jet engines... both for powered-landing and self-ferry capabilities

(It's a start... )

 

[zerofay32]

The problem with all spacecraft is the dangers of reentry. It's the one phase of space flight that if something goes wrong there is very little that can be done as far as abort possibilities. This is true will all manner of spacecraft. I don't believe that anyone can build a design that is 100% safe through an Earth reentry. It's just something that comes with the territory.

Zerofay32

 

[Thunder Chicken]

The problem with all spacecraft is the dangers of reentry. It's the one phase of space flight that if something goes wrong there is very little that can be done as far as abort possibilities. This is true will all manner of spacecraft. I don't believe that anyone can build a design that is 100% safe through an Earth reentry. It's just something that comes with the territory.

Zerofay32

True - I think the Soyuz system is about as safe as you can get for re-entry. As long as de-orbit burn goes OK, even if control goes down, you have a ballistic fail-safe mode that will still get you home.

I wonder if a winged spaceplane could be designed to be hands-free stable through re-entry should hydraulics or RCS fail. Spaceships 1 and 2 have the shuttle-cock wing, a good idea but requires moving parts and the thermal loads aren't near what orbital vehicles need to deal with. Maybe with additional forward thinking you could make the design 'stable' even after catastrophic loss of a lifting surface (maybe flying a lazy corkscrew path with stable AOA). A fat, stocky lifting body (like the HL-20, anyone?) with no wings to fall off might be the best bet.

If something goes wrong with pressurization and the suits and such, you're in trouble, but that is the case anywhere in the mission, for any ship.

 

[Linguofreak]

You've still got the problem of the spacecraft falling apart in an unpredictable manner after a heat shield failure, possibly exposing the crew to hot airstream.

I think a capsule design is optimal here simply because there are fewer events that can cause a heat shield failure. You don't have foam hitting the heat shield because the heat shield is nestled between the capsule and the upper stage of the launcher. The only event that can damage the heat shield is catastrophic failure of the upper stage (which is fairly likely to be fatal anyways), catastrophic failure of the capsule (which is fairly certain to be fatal anyways), or collision with orbital debris, which is a hazard for all spacecraft and can probably be mitigated by certain measures.

 

[SiberianTiger]

A quote from the mentioned report:

The first event with lethal potential was depressurization of the crew module, which started at or shortly after orbiter breakup.
The majority of the SCSIIT findings related to the first lethal event were connected to the operational incompatibilities of the advanced crew escape suit (ACES) with the orbiter. The launch and entry suit was added in response to the Challenger accident, rather than as a part of the original vehicle design. The ACES was the successor to that suit. The suit protects the crew in many scenarios; however, there are several areas where integration difficulties diminish the capability of the suit to protect the crew.
Integration issues include: the crew cannot keep their visors down throughout entry because doing so results in high oxygen concentrations in the cabin; gloves can inhibit the performance of nominal tasks;
and the cabin stow/deorbit preparation timeframe is so busy that sometimes crew members do not have enough time to complete suit-related steps prior to atmospheric entry.

Also:

The ascent and entry suit system provides protection from ground impact with a parachute system. The current parachute system requires manual action by a crew member to activate the opening sequence.

Not minding the incapability of anything to save the Columbia's crew during peak reentry heating phase, the highlighted points raise questions about applicability of the ACES suites for a shuttle's crew rescuing even in a less extreme catastrophic situations. I understand this is a risky business, but then, what's the point of wearing those orange suites at all?

 

[garyw]

True - I think the Soyuz system is about as safe as you can get for re-entry. As long as de-orbit burn goes OK, even if control goes down, you have a ballistic fail-safe mode that will still get you home.

Not always. There are always the dangers of parachute failure or seperation failure. As recent as Soyuz TMA-11 there were problems with the Soyuz coming in hatch first.


-----Post Added-----

A quote from the mentioned report:
Not minding the incapability of anything to save the Columbia's crew during peak reentry heating phase, the highlighted points raise questions about applicability of the ACES suites for a shuttle's crew rescuing even in a less extreme catastrophic situations. I understand this is a risky business, but then, what's the point of wearing those orange suites at all?

As I understand it, the ACES suit is not designed to save anyone during exposure to the mach 17+ forces of entry. It's not built for that. It can save astronaunts lives if the cabin depressuries or if there is a loss of control below about 200,000ft. From there they can bail out and use the parachute (if the orbiter is flying resonably level, if it's tumbling forget it). The ACES suit also has a bunch of other stuff for survival in cold water, etc, etc.

 

[SiberianTiger]

As I understand it, the ACES suit is not designed to save anyone during exposure to the mach 17+ forces of entry. It's not built for that. It can save astronaunts lives if the cabin depressuries or if there is a loss of control below about 200,000ft. From there they can bail out and use the parachute (if the orbiter is flying resonably level, if it's tumbling forget it). The ACES suit also has a bunch of other stuff for survival in cold water, etc, etc.

Please read my post carefully, I said there's no point in putting those suits on if the astronauts routinely don't close helmet visors or wear gloves during landing. Are you sure they would be able to attach those gloves quickly enough during a catastrophic depressurization at 200,000 ft?

 

[Coolhand]

I like the comparision to F1 cars that Thunder chicken made - drivers today routinely survive and walk away from what would have been fatal accidents back a few decades ago.

Perhaps in the coming decades an accident like columbias will be survivable, the tragic sacrifice of the crew and reports like this will contribute a lot to new developments, or at least the willingness to put the knowledge to use to improve the safety of future crews.

 

[silent_protagonist]

I think the problem with a Columbia-like situation still boils down to the heat shield, if that goes there's really not much you can do (unless you can figure out how to give each astronaut an individual backup heat shield, and figure out how to deploy them from the wreckage of breaking up spacecraft; might be theoretically possible but in reallity I doubt it would ever work). Therefore protecting the heat shield becomes the priority.

It seems to me that the two best ways to do this are to use a less fragile heat shield (i.e. a non-reusable ablative shield or some sort of advanced reusable shield), and to keep that heat shield snuggled safe inside the spacecraft for as long as possible (i.e. a capsule design, also has the advantage of allowing hands-free ballistic reentries). Of course, these are two of the design changes that people hate the most about Orion, but it makes sense.

 

[garyw]

Please read my post carefully, I said there's no point in putting those suits on if the astronauts routinely don't close helmet visors or wear gloves during landing. Are you sure they would be able to attach those gloves quickly enough during a catastrophic depressurization at 200,000 ft?

Not at all. This is why NASA has reminded all crews that gloves must be ON before EI. In the event of a depressuration all they have to do is close the visor. This can be done in the 20-30 seconds of useful consiciousness that they would have.

Many don't like the gloves though as they find them cumbersome to use. That's another issue.

 

[Thunder Chicken]

Wings to provide crossrange for the sake of reusabilty does not drive down the costs to get to orbit, at least on the scale of the STS, and the complexity and safety of the arrangement is another nail in the coffin. I think something on the scale of the HL-20 or Kliper coupled with a heavy lifter for cargo might have made the economic math come up better, and some of the structural / safety concerns go away in that arrangement. I am boggled (in a good way) that something as ambitious as the STS ever made it to the pad.

The cheapest and safest way to and from orbit is going to be in a capsule, and we need to get back into the business of flying rockets. We need to fly rockets until it becomes almost routine, like flying an airliner across country. We need to be able to access orbit 'at-will' if we are ever to do any serious exploration beyond orbit.

Soyuz is a nice mature technology for reliably getting to orbit - NASA needs to develop a similar basic way to orbit. They should develop something that they plan to fly to orbit for the next 50 years, with improvements and refinements.

I confess that I was initially disappointed and deeply skeptical of the Constellation/Ares program, but I am now converted (and I'm an engineer). I think the Ares rocket is the shortest path from here to a reliable safe ride to orbit. It takes advantage of a lot of existing technical capability and learning. The Ares 1X test launch makes 100% sense from an engineering development point of view, as they need flight data to validate the design tools for structural vibration and control analysis. It does not need to be identical to the final article, as the design will likely be modified for the Ares 1Y base on the 1X flight data and analysis.

 

[RocketMan_Len]

I wonder if a winged spaceplane could be designed to be hands-free stable through re-entry should hydraulics or RCS fail. Spaceships 1 and 2 have the shuttle-cock wing, a good idea but requires moving parts and the thermal loads aren't near what orbital vehicles need to deal with. Maybe with additional forward thinking you could make the design 'stable' even after catastrophic loss of a lifting surface (maybe flying a lazy corkscrew path with stable AOA). A fat, stocky lifting body (like the HL-20, anyone?) with no wings to fall off might be the best bet.

Safety and versatility are trade-offs in every field. The ultimate in safety is to not go.

Wings are not *inherently* useless - like every other part, they're structures that can be used. The problem with the Shuttle is that the wings are empty of anything save the landing gear. Add in fuel tanks and you get a benefit. Wings also mean the entry interface is higher in the atmosphere... so air density is lower and thermal load is reduced. Less thermal load means less overall stress on the airframe.

(Oh - out of curiosity, has there ever been an airplane that has been designed to be hands-free stable in all failure modes...?)

 

[Urwumpe]

(Oh - out of curiosity, has there ever been an airplane that has been designed to be hands-free stable in all failure modes...?)

Never. All failure modes already implies that there can be really bad situations - without a chance of automatic control working.

The UAVs are possibly the closest we got so far.

 

[garyw]

"all failure modes" also means it's not possible - Take the failure mode of a DC-10 losing an engine - how do you maintain stability with a huge chunk of the wing leading edge missing?

 

[Thunder Chicken]

"all failure modes" also means it's not possible - Take the failure mode of a DC-10 losing an engine - how do you maintain stability with a huge chunk of the wing leading edge missing?

In a DC-10, you don't. The DC-10 was not designed with the idea of an engine pylon failure being a plausible event, and so it was not designed specifically to survive that event.

What I am suggesting is asking the question during the design process: Given a potential failure mode for a conceptual design, how can I modify the design to render that failure mode harmless, or at least less harmful? The hardest part of engineering design is brainstorming ALL of the ways a design can fail and guarding against them, but it can be done if you think to ask that question.

In civil aircraft, the implicit assumption is that the airframe remains more or less intact. Failures like wings falling off and such are low probability events, so you design to avoid that failure, not to make it survivable.

Some combat aircraft were designed assuming the possibility of massive airframe damage and doing what was possible to make such events survivable (a good example is the A-10 Thunderbolt II). F1 cars are designed with the likelihood of a high-velocity crash in mind.

Judging the probability of a massive structural failure in a spacecraft (e.g. wing failure on the shuttle) is hard to do due to the complexity of the system and the harsh environments encountered. Loss of a shuttle wing or hydraulics during re-entry were evaluated as low-probability events, and so no fall-back mode was provided to moderate those failure modes.

Knowing what we know now, could we take a conceptual space-shuttle like design (STS version II, say) and design out those failures or modify the design to make those failures survivable? One answer might be to just bag the wings and go with capsules. But if the cross-range was really important and you just needed to have lifting surfaces, how would you design it to minimize risk?

 

[garyw]

Knowing what we know now, could we take a conceptual space-shuttle like design (STS version II, say) and design out those failures or modify the design to make those failures survivable?

Yes. Absolutey.

BUT!

Any new design will have it's own problems. Working out where those problems are and finding workarounds is key. For example, Looking at Oxygen tank layouts to improve surviability in the event of an Apollo 13 incident.

The problem is not in finding workarounds for the things we know about but the things we don't. You can run a vehicle fine for years and then some chain of events leads you to rethink the design. The problem is not in what we know and have learnt but in what we DON'T KNOW and in how much it will cost to refit the fixes into the design.

Putting this another way - How much do you spend to give the crew a 98% chance of survival? $2Billion? $3Billion? At what point do you say "Fly as is"? what if $2billion gives you 98.5% yet another $500 million pushes that to 99%? What if that change will take five years? Do you not fly for five years? What else will push the cost up in those years?

 

[Andy44]

Not minding the incapability of anything to save the Columbia's crew during peak reentry heating phase, the highlighted points raise questions about applicability of the ACES suites for a shuttle's crew rescuing even in a less extreme catastrophic situations. I understand this is a risky business, but then, what's the point of wearing those orange suites at all?

My cynical answer is that the suits are functionally almost pointless, and are mainly intended to give the public, and to a lesser extent, NASA personel, a "warm and fuzzy" feeling that something was done to make STS safer after Challenger. STS was designed to be a comfortable shirt-sleeve environment. The cockpit controls are not meant to used while wearing bulky gloves, nor is the life support system designed to operate when nobody is breathing the air (ie using closed visors).

The pole-bailout system is at least a small improvement over the alternative, which is to ditch at sea and hope you survive both the landing and the subsequent scramble out the hatch while the vehicle sinks into the sea. The astronauts call the ditching checklist "something to read while dying."

Could we build a safer, less-expensive space shuttle? Almost certainly. One thing you have to keep in mind is that STS is the first of its kind of system. Before STS-1, nobody had ever flown anything even remotely like it. We now have a quarter century of experience and over a hundred launches worth of data to put in our notebooks in the event we want to do it right. It would've been nice if Buran had developed a flight history as well, to add data from the perspective of a space shuttle with a different design philosophy, but if wishes were succesful launches...