Air France 447 - On topic only!

[Highkeas]

Oh, I know. I'm not even debating passenger aircraft should be built to withstand crashes. I was only comparing the F1 incident to the car crash tests posted by Gabriel and I was wondering why the latter is expected to be fatal (even if you excluded any debris invasion), while the former is expected to be survivable. I understand whatever the reason is, it can't really be applied to the big jets, in terms of physics and feasibility.

The level of acceleration a human body can withstand is dependent on exposure time - for example an ejection seat imposes 100+ g for a very short time (milliseconds) whereas a F-16 pilot can withstand around 6 g for tens of seconds when wearing an "anti-g" suit.

For the F-1 situation first the seat is securely fastened to the vehicle and the driver wears a 5-point seat belt. Plus race cars are designed to absorb energy during impact.

The seats for some flight attendants face backward and have 4 or 5 point seat belts to better withstand a crash (usually running off a runway or gear up landing). The rule for passenger seats was that they could withstand forward loads 9 g but I understand the newer rule requires 17 g.

[Joe H]

The level of acceleration a human body can withstand is dependent on exposure time - for example an ejection seat imposes 100+ g for a very short time (milliseconds) whereas a F-16 pilot can withstand around 6 g for tens of seconds when wearing an "anti-g" suit.

For the F-1 situation first the seat is securely fastened to the vehicle and the driver wears a 5-point seat belt. Plus race cars are designed to absorb energy during impact.

The seats for some flight attendants face backward and have 4 or 5 point seat belts to better withstand a crash (usually running off a runway or gear up landing). The rule for passenger seats was that they could withstand forward loads 9 g but I understand the newer rule requires 17 g.

Although is interesting, it wouldnt have made the slightest bit of difference in this case, and thats where the dollars count. ! . they could have had airbags surrounding the whole body, seats facing wherever, they would still be dead, the incident happened in the middle of the Atlantic ocean. It took more than the survival times in perfect conditions to reach or even find them. so it is acedemic. Evan is correct to a point. they cannot and will not build survivable aircraft in the event of a catastrophic accident or incident, they can certainly, and have, improved survival aspects, but only to a point. if we built them to be survivable, then they would not fly.. period..

[Evan]

So, how, with a fairly high downward velocity, do you hit in a 'fairly level' as in NOT nose down config ?

you have just recovered or are trying to from a huge stall ?

Possibly.

you were heading nose down then viewed your impact point and pulled up ?

Possibly.

You were in a flat flight but ina massive sink with no warning ?
or as I said you were in a major news nose up, tail first, Massive sink (STALL!) slam the tail first then body into the sea situation leading to the g forces measured...

I don't know about this one. To be falling out tail first you would have to be fully stalled or moving with considerable forward speed as well. Remember, the three clues to focus on here from wreckage compression are indications of a high vertical component, a slow forward speed AND a (more or less) wings level attitude. If they were fully stalled, that would mean a loss of roll stability. I tend to think they were either in some sort of early stage of stall or dive recovery (see Colgan) or were working the ECAM messages and had simply not noticed their alt or v/s indications until it was too late (see my thrust-lock theory a ways back).

[Joe H]

Possibly.



Possibly.

You were in a flat flight but ina massive sink with no warning ?
or as I said you were in a major news nose up, tail first, Massive sink (STALL!) slam the tail first then body into the sea situation leading to the g forces measured...

I don't know about this one. To be falling out tail first you would have to be fully stalled or moving with considerable forward speed as well. Remember, the three clues to focus on here from wreckage compression are indications of a high vertical component, a slow forward speed AND a (more or less) wings level attitude. If they were fully stalled, that would mean a loss of roll stability. I tend to think they were either in some sort of early stage of stall or dive recovery (see Colgan) or were working the ECAM messages and had simply not noticed their alt or v/s indications until it was too late (see my thrust-lock theory a ways back).[/quote]

Hi Evan. I mostly agree . - but that does not fit with the wreckage / debis field.
Remember vertical speed - down - is a component of the vector.....


So a high VS with low Foreward speed is similar... - in the actual compression of metal or composite materials.

However, here we have a large debris field which WOULD tend to indicate a fairly high forward vector. one cannot have one without the other.

[Joe H]

Possibly.



Possibly.

You were in a flat flight but ina massive sink with no warning ?
or as I said you were in a major news nose up, tail first, Massive sink (STALL!) slam the tail first then body into the sea situation leading to the g forces measured...

I don't know about this one. To be falling out tail first you would have to )
be fully stalled or moving with considerable forward speed as well. Remember, the three clues to focus on here from wreckage compression are indications of a high vertical component, a slow forward speed AND a (more or less) wings level attitude. If they were fully stalled, that would mean a loss of roll stability. I tend to think they were either in some sort of early stage of stall or dive recovery (see Colgan) or were working the ECAM messages and had simply not noticed their alt or v/s indications until it was too late (see my thrust-lock theory a ways back).

Look at the Debris field, !. you cannot have high VS and LOW Foreward S. not to produce that... so the vector is a combination of the two. To me, they simply had no idea at all as to speed or altitude - they spied the oggin, they pulled up in desperation. probably at around 250 - 300 kph forward... and at the bottom of a sinusoidal sink curve.

[guamainiac]

At that hour and with the rapid degradation of flight I don't think they ever saw the ocean coming up.

[Highkeas]

Look at the Debris field, !. you cannot have high VS and LOW Foreward S. not to produce that... .................

Are you talking about the shape of the debis field?
Assuming that there was some forward velocity of the aircraft and its result debris, objects with a high W/CdA will travel further though the water than objects with a low W/CdA. This produces the oval shaped debris field. Ocean currents will play a part as the objects sink. My bet is that the detailed debris mapping will bear this out. I'm sure preliminary mapping exists but is not public yet to my knowledge.

[Black Ram]

Thanks Highkeas. I suppose restraints have a lot to do with survivability. I remember a similar F1 incident in 1994 that resulted in instant death due to sustained neck injuries. The restraints back then were less advanced and didn't have the HANS device. Not that any of this is practical on a passenger jet.
When it comes to internal organ damage in a crash, it is absorption of energy that counts, as you mentioned. Does that mean Kubica's car absorbed so much energy, that the level of deceleration he sustained from hitting the wall at 180mph was substantially less than what one would have experienced in the Smart or the small Opel from Gabriel's video, when hitting the wall at 60mph?
So about AF447, is there an estimate for the deceleration sustained during impact? Is it possible anyone could have survived the impact? I read somewhere that supposedly some passengers were cut in half by their seatbelt - is that possible? I would have expected seats to tear from the floor.
All this despite Joe H being right it would not have mattered.

[Fear_of_Flying]

It would be nice to think that airline safety is continually being developed on a number of levels. First and foremost, of course, is to avoid accidents at all costs (figuratively speaking). Next, to make sure that as many people as possible survive "survivable" accidents. This seems to be the emphasis of some of the more recent innovations, such as airbags. Finally, to look towards making some "non-survivable" airline crashes survivable. I can see how this might seem more like science fiction in crashes where the plane and passengers end up in a million pieces on a mountainside. But perhaps in an accident where the bodies are intact, but fatally injured by internal trauma including closed head injuries, spinal compression, and so forth, it is a more realistic possibility. It is, after all, a straightforward problem to put to a physicist: how can we reduce the amount of energy absorbed by a passenger in certain types of crashes? I seem to remember designing a container that would keep an egg from breaking when dropped from a rooftop when I was in 7th grade. Same basic principle, right? One hates to simply dismiss the possibility of surviving certain types of crashes as unfeasible: today, maybe, with the technology we have, but 10 or 20 years from now? Hopefully there is research being done in that area. We have, as it has been pointed out, certainly made incredible advances in auto racing crashes.

[Gabriel]

I've seen that video and it makes me wonder why sometimes more severe collisions are easily survivable. For example, a few years back Formula 1 driver Robert Kubica smashed his Sauber-BMW also into a concrete wall, nearly straight into it while doing over 180mph. Telemetry revealed he sustained a peak 75G. This is what it looked like:

http://www.youtube.com/watch?v=MZqBbLyp6oo

He didn't even spend the night at the hospital. Makes me wonder if the type of restraint, the seating position, and one's physical condition have anything to do with it.

I don't think that that accident was so severe as you say.
I mean, wow, what a crash, but it's not a crash at 180 mph straight into a concrete wall.

In that video:

At the moment he losses control of the car, he surely starts to loose speed.
At 0:16 he has a small almost tangent crash against a wall. Some energy is lost there, and the car has to keep loosing some speed as it skids, until
At 0:17 the big crash happens. The view of the impact is blocked, but
At 0:18 you see it emerge behind the sign, still with a lot of speed (and hence still with a lot of energy.
At 0:19 you see spinning (that takes energy too)
At 0:20 it tumbles and rolls, loosing a lot of fuel and parts (more energy there), and it still keeps skidding at a somehow high speed.
At 0:22 it crashes against the opposite guard-rail. While this crash was no doubt much less severe than the previous one, it still takes a good bunch of energy with it, and then it still has some energy to skid a few dozens of feet (and 5 seconds) more.

All in all, 10 seconds happened between the big impact and the full stop, 10 seconds full of energy-consuming events (skidding, spinning, tumbling, rolling, loosing fuel and parts, and another crash against a guard rail).

Again, while a lot of the energy was consumed very quickly in the big impact, it's not nearly as crashing directly into the concrete wall. A big component of the speed vector was parallel to that wall and that component was almost intact after the big crash.

On the other hand, in the Smart and Opel crashes they do indeed hit the wall almost directly into it. There is only a small component of the speed vector parallel to the wall and you can see that, after the crash, the cars travel very slowly just a few meters in a very low "energy consumption" event (the just run and skid, no second crashes, no spinning tumbling, rolling or loosing parts).

I'm not saying that the accelerations suffered by Kubica were less than those that an eventual driver of the Smart and Opel would have sustained. Just trying to explain that the F1 crash was awfully violent, but just not as much as one might initially think.

[Black Ram]

Pretty impressive, Gabriel. Turns out Kubica clipped the first obstacle at over 180mph, but somehow shaved off 40mph right after that, before getting airborne. After a few seconds in the air, with no chance to steer or brake further he hits the wall at over 140mph. However, the angle he hits it at is about the same as the crash tests in your video (the concrete is angled and not completely perpendicular to the trajectory). Also, the 75G is correct, but I don't know the time exposure. Furthermore, he not only survived, but got away with a concussion and a sprained ankle.
It really seems the key to surviving such a crash and avoiding the internal injuries is, as you mentioned, absorption of energy and maintaining velocity after impact, thus reducing the level of deceleration. It seems these cars are made to bounce off and keep a lot of the inertia, while shedding debris, which also takes away some of the energy. But the Smart and the Opel also could have kept going over the concrete wall, as opposed to the abrupt stop. Maybe there is a design limitation due to their small size. So maybe it is possible such a crash in those small cars subjects one to higher levels of deceleration than Kubica?
Finally, if Kubica's crash wasn't at full speed, the other F1 incident I referenced was. In 1994 Roland Ratzenberger lost his front wing on the fastest stretch of the circuit, becoming partially airborne and hitting the concrete at 195mph. Not only was the impact at top speed, but also the concrete wall was a lot taller and there was no jumping it (still the car bounced off and kept going along the wall, after the COMPOSITE bodywork absorbed a lot of energy). Sadly, Ratzenberger was killed (probably instantly, though he was pronounced dead at the hospital). However, it was a basal skull fracture that killed him. There was no internal organ damage and one might speculate that if the restraints then were as advanced as today's and complete with a HANS, the outcome may not have been fatal.
Here is Ratzenberger's accident if anyone wants o see it (some may find it disturbing):

http://www.youtube.com/watch?v=AkneSUex74c

[Black Ram]

Fear of Flying, I even think a lot of progress has been made. There are scenarios which completely overpower us, like high-speed nose-first crashes. But there are a lot of cases of no fatality in more survivable scenarios - the Hudson water landing, AF358, the SAS MD-80 which made an emergency landing on snow/ice after losing both engines.
Unfortunately given the circumstances of AF447, hitting stormy water with a high sink rate and some forward movement, I think it is most likely they went through immense deceleration, possibly coming to a halt in a few seconds.

[Fear_of_Flying]

I guess my point was more that I assume that for any technology-based industry, research and development has short-term, intermediate-term, and long-term divisions. So if you are an engineer for Ford, you might be improving a current model of vehicle, working towards the next generation of vehicle, or working more theoretically with prototypes to develop future technologies. I imagine the same is true for airline safety, so maybe there are researchers already looking ahead to making high energy crashes survivable.

In any event, I looked at the NASA Aviation Safety Program Fact Sheet, and I found that there are three main areas that they focus on:

• Accident Prevention
• Accident Mitigation
• Aviation System Monitoring and Modeling

They address these areas using 8 technology strategies:

1. Make every flight the equivalent of clear-day operations.
2. Bring intelligent weather decision-making tools, including worldwide real-time moving map displays, to every cockpit
3. Eliminate severe turbulence as an aviation hazard
4. Continuously track, diagnose and restore the health of on-board systems, leading to self-healing and "refuse to crash" aircraft
5. Improve human/machine integration in design, operations and maintenance
6. Monitor and assess all data from every flight for both known and unknown issues
7. Increase survivability when accidents do occur
8. Anticipate and prepare for future issues as the aviation system evolves

To me, it is interesting how pertinent each of these strategies is to AF447. Advances in just about any one of them, at least 1-6, could have potentially helped prevent the accident.

[Evan]

To me, it is interesting how pertinent each of these strategies is to AF447. Advances in just about any one of them, at least 1-6, could have potentially helped prevent the accident.

Number 7 is definitely not pertinent to AF447, which is the point I was trying to make. Interesting list though. All but #7 are preventative measures.

[Jpmkam]

Recovery ship now on-site.

SAO PAULO - A French investigation agency says a recovery ship has arrived in the area where an Air France jet crashed off Brazil's coast about two years ago.

All 228 people aboard the Airbus aircraft were killed when Air France Flight 447 from Rio de Janeiro to Paris crashed June 1, 2009. The cause of the crash remains unclear.

The agency said in a statement that the private cable-laying ship Ile de Sein arrived at the wreckage spot early Tuesday morning and that an underwater robot will be used to help locate and recover the flight recorders in the jetliner's rear section.

Experts have said that without retrieving the voice and data recorders there would be almost no chance of determining what caused the crash — the worst disaster in Air France's history.