Polish President and wife killed in Tu-154 crash

[ACheck]

I'm amazed that this thread is 115 pages long. Isn't time to end it?

[3WE]

I'm amazed that this thread is 115 pages long. Isn't time to end it?

I guess you are one of the conspirators and are getting nervous?

No, the thread shall not be closed, the truth must continue to be spoken.

[Northwester]

Binienda did a crash simulation using exact parameters as included in the official reports. Horizontal speed, vertical speed, angle of impact, roll, ground properties, etc. The plane did not fall apart.

Remember, this is the type of work Binienda does for NASA.

http://www.youtube.com/watch?feature...&v=aDBWuqduS2U#!

[Gabriel]

Binienda did a crash simulation using exact parameters as included in the official reports. Horizontal speed, vertical speed, angle of impact, roll, ground properties, etc. The plane did not fall apart.

Remember, this is the type of work Binienda does for NASA.

http://www.youtube.com/watch?feature...&v=aDBWuqduS2U#!

Any remaining credibility that Binienda had with me has just gone down the drain.

[Northwester]

Any remaining credibility that Binienda had with me has just gone down the drain.

Now you are loosing your credibility. What position are you in to question his work?

[SYDCBRWOD]

From that dodgy video - the outline of the aircraft seems to be hitting the ground inverted yeah?

1. So it has rolled inverted in only a few seconds at most - maybe less than a second, yet suddenly arrested its roll so that in its final second or so it was no longer rolling (the video simulation didn't seem to show any). So we are to believe that the airliner has aerobatic control abilities like that of a Pitts Special.

2. The tailplane did not snap off or cause the failure of the aft fuselage after hitting the ground as a second reason.

3. The speaker should cut down on the amount of alcohol before addressing an audience. I don't know about you, but I couldn't understand him.*

There are probably more that Gabriel was thinking of but they are the two that came immediately to mind.


* For those without a sense of humour - yes, I know he was speaking Polish.

[Gabriel]

Now you are loosing your credibility. What position are you in to question his work?

First of all, I care very little of what you think of my credibility, very much like Binienda cares very little about what I think of his credibility.

But I am just a screen name in an internet forum and he is a Proffessor that works for the NASA and that is questioning publicly two official reports and supporting a conspiracy theory. So his credibility is much more important than mine for the sake of this accident.

Second, his credibility was already low by stating things like a plane can't climb with a fraction of its wing missing, and thus the Russians are lying and inventing a physically impossible climb to fit the wing between the airplane and the ground.

Third, he is using 100% of data from the official report, except that the data stopped recording when the plane was still climbing and rolling past about 90°.
As far as I remember (and I can be wrong on this), there is no info in the official report about the speed, vertical speed or pitch angle at which the airplane impacted the ground.
I do remember, however, that the report concluded, from ground scar marks, that the airplane hit the ground with a roll angle of 200 to 210° and still rolling, while this animation clearly shows the plane hitting inverted with wings level and not rolling.

Fourth there is some common sense combined with engineering intuition here. I know this is fuzzy, but for me seeing this simulation is equivalent to seeing a simulation of an egg being strongly hit with a big, heavy hammer where the egg wobbles but doesn't brake. Would you believe in Einstein himself showing this and saying "this is a proof that you can't brake an egg with a hammer"?

Fifth, let's test that common sense and engineering intuition.
Since I don't have the data that doesn't exist in the official report and I don't know what numbers Binienda used for this simulation, let me make a wild guess that the plane was flying at 300km/h (the approximated speed it had along the approach, and while it was reduced a bit in the latter stages remember that full thrust was added shortly before the virtual hit with the birch) and impacted the ground with a trajectory of 10° of descent slope. (the approximated angle between the top of the cockpit and the top of the fin is about 11°, and you can see these two pints contacting the ground almost simultaneously, and while this is an indication of pitch and not trajectory, I'll use it in the lack of a better number).
That will make for a vertical speed of 2800ft/min.
Now, the regs (at least the USA and UE ones) require that an airplane resists a landing at 640ft/min, and that's when landing on tires, shock absorbers and a structure specifically designed to resist these loads attached to the wing spars and near the CG of the plane.
Why would someone put a resistant aka heavy enough structure to withstand a vertical speed of 4 times that much supported only by the upper frame of the windshield and the fin? It doesn't make any sense.
Not only that, the report says that the scars in the ground reached depth of 0.5m and, from the simulation, it looks like the structure at the point of impact deforms just a few inches, let's say another 0.25m (10 inches).
That will make for an average acceleration of 14g.
Now, the REGs require for transport category airplanes 2.5g with an ultimate load of 1.5 times that much: 3.75g.
And that's when that load is borne by the whole wing surface that is specifically designed to do that.
How is that the airplane resists an acceleration of 4 times that much supported by the top of the cockpit and the fin?
And then, the fin (or supporting structure) not being torn off by the horizontal forces is simply unbelievable. The fin, together with the top of the cockpit, is slowing down the whole airplane. A piece that is designed to withstand, in the horizontal direction, just the aerodynamic drag of itself and the stabilizer, is now slowing the whole mass of the plane and doing so at a high acceleration rate? Yes, sure.
The fin of AF was not strong enough to slow down itself, but this fin is slowing down the whole airplane.
And the TU-254 is a pretty efficient airplane in terms of weight, with a MTOW:empty wight ratio of 2:1, which is a world class figure, comparable to that of the 787. So it's not that they had a lot of structure beyond what is needed to meet the standards.

So it's impossible to brake an egg with a hammer?
Nope, sorry. Not in my world.

[Northwester]

Gabriel,
First of all, I respect your opinions in the area where you know what you are talking about. You are one of the few people here who have a legitimate reason to be a part of the discussion because you refrain from personal attacks and talk about substance. Binienda is one of the top experts in the US (and probably in the world) in the area of high velocity impacts and this is where I respect his opinions. When he is talking about aerodynamics he might be quoting someone else, and might not be explaining the whole story. You know very well that when the plane reaches a roll of about 50 deg, in most cases it cannot gain any altitude except from the inertia. And at 90 deg. roll doesn't matter how many g's the pilot can pull (could be 100), the vector is pointing down, and in our case with very low inertia, the plane is not flying up.

But back to the real issue. There are a lot of contradictory data in both reports about the flight parameters in the last phase of the flight. The Russians are saying that at the first contact with the ground the plane was at 200 - 210 deg. roll. But at the same time they are saying that the plane first touched the ground with its left wing and stabilizer leaving those two marks on the ground. Imagine the plane at 200 deg. roll and touching the ground with the left wing and stabilizer! Pure nonsense. On top of that if you check the distance between the left wing and stabilizer and the distance between the marks on the ground, they don't match (see the pic). In the Polish report they maintain that to roll at the impact was about 150 -160 deg. How do they know it? I don't know. It is a calculated value based on some assumptions. If you look at the values they use for the position of the plane in the time progression and corresponding roll, you see that according to them the roll speed (deg/sec) varies wildly, as though the plane was a little toy with no mass and inertia. Maybe they based it on the tree damage. But again their values don't make sense, and it does not matter if this is an official report or not, the numbers have to work.

The data stopped being recorded at the moment when the FMS froze, about half a second before the impact. The highest roll recorded is 62.5 deg. The maximum range for the sensor reading the roll is about 80 deg. Besides that we have statements from the witnesses that contradict the official roll values. The bus driver stated that saw the plane flying leaning to the left, with the wheels pointing down where according to the reports the roll would have to have been at least 90 deg. Another witness said: "I looked through the window and saw that the plane was falling - it was leaning to the left, nose down. It was flying in the fog, slicing the tops of the trees. Before the impact the tail fell off. Then it fell in the woods. I heard a strong explosion - and a blast to the top of the trees".

Binienda got the basic parameters from the QAR box, the air speed of 270 km/h and the vertical speed of 10 m/s. He used these values to show that in an inverted fall with such horiz. and vert. speeds the plane CANNOT SUSTAIN FRAGMENTATION that occurred in Smolensk (see second pic). There was a significant damage to the fin and to the front of the plane, but the fuselage remained in one piece. He repeated the simulation with 20 m/s vert. speed and then the fin broke off at impact and the damage to the front was more severe, but still nothing close to the fragmentation that we can see on that pic. That was the whole purpose of his simulation. As of now no one knows exactly what was the position of the plane at the impact, but the inverted impact into the soft, wet ground, with the above shown speeds, could not have resulted in the total fragmentation of the plane. That's what he is showing. He is showing that what's presented in the report cannot be real without some other factors at play.

[Gabriel]

Hey, Northwester, I am not defending the Russian report.
But it's not me who said:

Binienda did a crash simulation using exact parameters as included in the official reports. Horizontal speed, vertical speed, angle of impact, roll,

Binienda [...] air speed of 270 km/h and the vertical speed of 10 m/s. He used these values to show that in an inverted fall with such horiz. and vert. speeds the plane CANNOT SUSTAIN FRAGMENTATION that occurred in Smolensk (see second pic). There was a significant damage to the fin and to the front of the plane, but the fuselage remained in one piece. He repeated the simulation with 20 m/s vert. speed and then the fin broke off at impact and the damage to the front was more severe, but still nothing close to the fragmentation that we can see on that pic [...].
the inverted impact into the soft, wet ground, with the above shown speeds, could not have resulted in the total fragmentation of the plane. That's what he is showing. He is showing that what's presented in the report cannot be real without some other factors at play.

Let me see:
270km/h = 2.5 times a typical highway limit.
10 m/s = 2000 ft/min = 3 times what's required for the landing gear.
20 m/s = 4000 ft/min = 6 times what's required for the landing gear.
The plane is made of Al and has en efficient and lightweight structure, enough to have a MTOW:empty weight of 2:1.
It hits inverted in a low pitch attitude (nose first) and it basically remains in one piece, or at most it looses the fin and some "more severe damage" to the front, with no significant fragmentation.

I am sorry. That's beyond my ability to believe. I am seeing Prof Binienda showing a simulation of an egg being hit with a big, heavy hammer, and the egg not breaking. I don't care who says so. I've seen enough broken eggs to know it's not true.

Maybe I am wrong, but I cannot believe it.

Not until I see a real Tu-254 doing that or until I understand exactly the calculation made, and that means that I can replicate it and reach the same result. And I very much doubt any of the too will happen.

You know very well that when the plane reaches a roll of about 50 deg, in most cases it cannot gain any altitude except from the inertia. And at 90 deg. roll doesn't matter how many g's the pilot can pull (could be 100), the vector is pointing down, and in our case with very low inertia, the plane is not flying up.

Would you please stop this bullshit?

I've just re-done the simulation of the airplane rolling inverted with the parameters from this flight. See the attachment.

I want you to understand what I did:
- I set an initial condition of -8 m/s = -1580ft/min of vertical speed and 270 km/h of airspeed (I assumed no wind that that this speed remains constant along all the event, which will not be true but will not have a big impact on the outcome anyway).
- I made a load factor vs time that closely matches that on the report. Interesting to note that, unlike the previous simulation where the load factor was 2 and 1.8 for a big part of the simulation, in this case it topped at 1.35 for just 1.5 seconds and was at or below 1.1 most of the time. So it's nothing spectacular.
- I modeled a roll dynamics (rolling moment, moment of inertia, and damping moment) to obtain a bank vs time that closely matches that on the report. Note that if the bank vs time in the report would have been ridiculous I could not have modeled it because the only thing that I modeled was the roll acceleration (not even the roll rate) and I used the standard model of a constant rolling moment, a constant moment of inertia, and a damping moment proportional to the roll rate. I let Newton derive the roll rate and bank angle from that.
- From the above data, I let Newton compute the accelerations in the ground-referenced x, y and z axis, and from that acceleration and the initial speed and vertical speed I integrated the speed in x, y and z every 0.1s, and from the initial position and those speeds I integrated the positions in x, y and z every 0.1 seconds.
- So basically, the vertical speed, altitude and ground track were completely left to Newton.

Not surprisingly, I obtained an altitude, vertical speed, and trajectory profile that closely match that in the report.

Now, if I was a Russian wanting to make a fake report, I'd do just tat: ensure that the data is consistent. So my results don't prove the truth of the Russian report (at least it proves that, if it's invented, they were careful enough to make it self-consistent).

But what it does prove is that it is perfectly possible for the plane with this "low inertia" to climb after loosing the wing (load factors of between 1.35 and 1.1 are perfectly attainable with 10% of the wing missing at any reasonable speed not too close to stall) and still be gaining altitude when the bank angle passes through 90°. And with enough room between the fuselage and the ground to fit a wing with room to spare, especially a shortened wing.

[Northwester]

But what it does prove is that it is perfectly possible for the plane with this "low inertia" to climb after loosing the wing (load factors of between 1.35 and 1.1 are perfectly attainable with 10% of the wing missing at any reasonable speed not too close to stall) and still be gaining altitude when the bank angle passes through 90°. And with enough room between the fuselage and the ground to fit a wing with room to spare, especially a shortened wing.

I have to apologize for the not entirely accurate statement about Binienda's simulation. He used the vert. and horiz. speed, and placed the plane in an inverted position. It does not seem that he applied the roll as shown in the reports. But please keep in mind that he did not assume any particular results. He, and his team, spent several months building the virtual model of the plane, modeled all structural elements, applied structural properties to them, applied ground properties, and let the computers do their work. He might be not 100% accurate, but based on his previous simulations where they compared the simulations to actual real life experiments, he is probably not that much off. For sure his simulation is much closer to reality than what we are shown in Smolensk, with the assumption that the causes of the crash are as explained in the reports.

Your hammer and egg metaphor is compelling with some conditions. The ground was soft and wet, and the plane did not fall straight down. So if you wrap the hammer with foam padding, place the egg in a sandbox, and hit it from the side, you might get something that is closer to reality. The egg might still crack, but maybe it will not be smashed entirely to pieces.

I took your beautiful graph (sorry, did not ask you for a permission first) and superimposed some values from the Polish report. The bank angle values are a bit awkward, they almost hint on two separate curves. Some of the vert. speed values are way off. But the most interesting is the altitude part. The report values are much lower where the plane was hitting the trees (according to the report), your curve is almost exactly where the independent researchers are placing it, slightly above or just skimming the trees. Would it be possible for you to adjust the alt. curve to match the green points and see if everything else still works?

[Gabriel]

Northwester,

What I've said about the egg and the hammer was in no way intended to be an analogy with the accident. It was just to explain the particular and subjective emotion that I feel when I see that animation: disbelief, as if I was looking a simulation of an egg being hit by a hammer and not breaking.

I'd like to see the simulations of the Turkish 737 that crashed short of Amsterdam. It stalled and touched down wings level and nose high. It had a much lesser sink rate, a much slower forward speed (i.e. much less energy to dissipate), the landing gear down, and the wing under the fuselage (so the wing could cushion the fusealge instead of the fuselage absorbing the impact from the wing above). It also crashed on soft mud. It broke in 3 parts IIRC.

The recent crash of another 737 short of the runway on water. We don't now much about it, but probably both the forward speed and (especially) the sink rate were much lower than the Tu. And the water cushions more than mud. It broke in two.

Or the 727 in the desert in Mexico, the test crash. The airplane simply made a hard landing. Nose high, normal approach sped, landing gear down, a sink rate much higher than the limit but much much below that of the Tu. It broke in 2.

The fact that the plane crashed with forward speed and not just fell down flat is worse, not better. If you just drop the plane in a given attitude so it crashes with a given vertical speed, the structure will compress and deform as needed to stop the vertical motion. If you add forward speed to that, then you must add the friction force too, that can be as much as the vertical forces sustained during the crash. And then, when the structure stopped the vertical motion, you have a deformed, weakened, broken structure to keep coping with the friction forces, an uneven terrain (even if mainly flat, imagine just something the size of a speed bump at 270 km/h), and whatever dynamics appear after that (tumbling, rolling, cartwheeling, etc.).

Regarding my graph, thanks for superimposing some points to it. What was the source of said points?

I am amazed. The matching is closer than I had anticipated. There are differences, of course.

Regarding the vertical speed and altitude, they are integrated from the acceleration. The original load factor plot is a mess, with a lot of peeks and valleys and lot of variation. So I did just a rough smoothed approximation.

Integration magnifies the errors. So if the load factor has errors, the vertical speed will have more, and the altitude even more (because it's the integration of an integration).

Regarding bank, I expected this difference in the latest stages of the roll. The model that I used, as I said previously, is the standard model for roll. This model disregards lift an just adds a constant rolling moment caused by the ailerons to the damping moment that appears as the airplane rolls (which is the one that makes the plane stop rolling when you center the ailerons, instead of keep rolling). In this case, however, I can make no assumption about the ailerons (we don't know if they were even operable after the supposed crash with the birch). The rolling moment will be the result of the asymmetric lift due to the asymmetry in the wing (only one was broken). Unlike the standard model, that moment WILL be a direct function of lift. No lift = no asymmetric lift = no rolling moment, high lift = high asymmetric lift = high rolling moment. That could be the reason why in the points that you superimposed (and that they don't match the page 156 of the Russian report that I took as basis) the rolling rate slows down when the airplane rolls inverted and the load factor is highly reduced.

Regarding your request of adjusting (reducing) the altitude, it's very easy to do but I don't feel like doing it because it's of little value. The result in the altitude plot will be just an offset down the same graph, and all the other graphs will remain unchanged. You can easily visualize it. Just subtract the desired adjustment to all the points of the graph.

In any event, as I've said, the Excel spreadsheet is available to anyone who wants it. Just PM me your e-mail and I'll send it to you, then you can play with it by yourself.

[guamainiac]

Huh? Oh! Going back to my nap. Wake me when there is some conclusion OK?

[3WE]

Huh? Oh! Going back to my nap. Wake me when there is some conclusion OK?

Conclusion: Lee Harvey Oswald acted alone.

[Gabriel]

Conclusion: Lee Harvey Oswald acted alone.

And fired a single shot: the magic bullet.

[Northwester]

An interesting news release today. A statement from FSB relating to the CVR:

"In recordings, included in given files, one could not detect credible signs of the continuity of the recording in sections separated by the known sections with signs of alterations. Because the original phonograms or other copies of the original recordings were not supplied, establishing if the recordings in the supplied flies were done continuously turned out to be impossible [...]. It is impossible to exclude the introduction of alterations to the recording, done digitally."

The statement is a part of an analysis prepared by experts of FSB Criminal Institute Valeriy Didenko, Maxym Ivanov, and Alexandr Ivanov on June 15th, 2010, file # 30-F/10.