PPRuNe Forums


Accidents and Close Calls Discussion on accidents, close calls, and other unplanned aviation events, so we can learn from them, and be better pilots ourselves.

Reply
 
Thread Tools Search this Thread
Old 14th Jan 2018, 19:38   #401 (permalink)


Probationary PPRuNer
 
Join Date: Dec 2013
Location: Norfolk
Age: 61
Posts: 2
Concours77

Fair points, but the aircraft did strike the ground 380 feet past the embankment and back of an envelope calculations are sufficient to prove that this is possible and reasonable without invoking some violation of the laws of physics. The aircraft was in a relatively shallow descent through most of the curved flight path and it was only after the angle of the wings exceeded 60° that the wings stalled and gravity became the predominant force. The rate of descent was not constant.

Of course the aircraft was in nothing like free fall...

So 100 feet past the embankment in a vacuum, maybe.
G0ULI is offline   Reply With Quote
Old 14th Jan 2018, 20:29   #402 (permalink)
 
Join Date: Nov 2016
Location: Lakeside
Posts: 263
Quote:
Originally Posted by G0ULI View Post
Concours77

Fair points, but the aircraft did strike the ground 380 feet past the embankment and back of an envelope calculations are sufficient to prove that this is possible and reasonable without invoking some violation of the laws of physics. The aircraft was in a relatively shallow descent through most of the curved flight path and it was only after the angle of the wings exceeded 60° that the wings stalled and gravity became the predominant force. The rate of descent was not constant.

Of course the aircraft was in nothing like free fall...

So 100 feet past the embankment in a vacuum, maybe.



That is what I wanted to establish, that the impact evidence is as observed. The flight path, not so much.

What do you make of the Lockheed experiment regarding possible conflict of expected airflows over the aileron, as influenced by the wing flap, without its jackscrew?

What it says to me is quite interesting. Why remove the jackscrew? At first blush, to simulate a deformation of the wing flap that affected roll? We must assume they don’t care a fig about what happened post impact, so is their surmise that the flap jumped its canoes inflight?

Wouldn't The consideration of drag embellish my point, and serve to conflict with a greater distance from first impact?

Interesting?
Concours77 is offline   Reply With Quote
Old 15th Jan 2018, 01:01   #403 (permalink)


Probationary PPRuNer
 
Join Date: Dec 2013
Location: Norfolk
Age: 61
Posts: 2
I think the tests were undertaken to eliminate fouling between the flaps and the ailerons as a possible cause of the accident. The investigators needed to consider all possible causes of the uncontrolled bank, in addition to an aileron roll, an asymmetrical extention or retraction of the flaps, could and probably would have produced forces that could not be controlled by the ailerons. It has happened on other aircraft designs. I well remember my flying instructor insisting that a visual check to both sides of the aircraft be made whenever the flap lever was operated. One of the things I used to regularly forget, especially selecting final stage flap just before landing. I knew the flaps on either side were mechanically connected, what could go wrong? Well they were connected with a joint that was capable of sliding and rotating. Copies of accident reports in the office soon persuaded me that it was a vital safety precaution to carry out a visual check in a small aircraft. Larger aircraft are not much different, things just happen at a different pace and control surfaces may not be visible from the cockpit, so you have to rely on other telltales.

So you are correct when you say that a flap fault could have produced similar results, but the solution is to return the flap lever to the position it was in before the flight upset started in order to restore a balance. Takeoff flap settings are usually only a few degrees and designed to increase the effective wing area while keeping drag at a low level. If one of the flaps had dropped completely, commanding full flaps would have rebalanced the aircraft to a level attitude and given the impressive power to weight ratio of the Electra, it should have been capable of maintaining altitude even with full flaps set.

Witness marks on recovered components suggested that takeoff flap was set and maintained throughout the flight and that the flaps were not set asymmetrically before impact.

All the additional tests were conducted to eliminate other possible causes before settling on a break in the aileron control cables as the cause of the accident. Of course the root cause was the incorrect actions of the maintenance engineers on the ground.

Were there indications of a potential problem during the takeoff run?

The rolling takeoff is significant in that it is usual to stop, verify runway heading and compasses are aligned and that all controls are free and move to their full extent. There is mention in the report that a test pilot felt virtually no change in the feel of the controls when the aileron cable was deliberately severed on a test rig.

If the full and free control movements were conducted on, or just coming off the stand, perhaps the aileron cable separated during this test. The trip along the taxiway to the runway would not have involved movement of the ailerons, just toe brakes, throttles and rudder.

The first indication that the pilots might have had that something was not quite right would probably been just before V2 when they were already committed to takeoff. Up until that point, the rudder would have been the sole directional control in use. Perhaps the aircraft did not respond correctly at that point to aileron input to counter a slight crosswind. At that point it could have been dismissed as slop in the control rigging, since the aircraft had a history of poor aileron response. It was only once the right bank on departure was initiated and could not be countered, that it became obvious that this was a critical failure in the control system. The accident report suggests that the pilots would not have been aware of the cable disconnect through any difference in feeling of the control yokes, only that the aircraft failed to respond to aileron control inputs.

The general rule is that once an aircraft has reached flying speed it is safer to lift off and try and solve problems once airborne than to attempt to stop. This was one of the exceptions to that rule.

The rolling takeoff may have contributed in some way to the accident, but that would rely on a visual check of the aileron movements being conducted from the cockpit and the lack of proper movement being noticed. I suspect that wing visibility from the cockpit is very limited if possible at all, like most larger aircraft.
G0ULI is offline   Reply With Quote
Old 15th Jan 2018, 01:57   #404 (permalink)
 
Join Date: Mar 2005
Location: N/A
Posts: 1,966
Quote:
Similar to the DC10, the pilot’s were along as pax.
In the DC-10 case that is how it ended up, but not how it began. The crew actually crashed an aircraft that was perfectly capable of flight, but aircraft design, coupled with crew training and procedures brought them undone.

When the engine departed it destroyed the hydraulic lines responsible for keeping the slats on the left wing open. With no hydraulic pressure the slats on the left wing retracted under imposed aerodynamic forces, hence increasing the stall speed for that wing.

Upon engine failure the crew reduced speed to the trained and procedurally required engine out climb speed. The left wing stalled in the process, and the aircraft rolled uncontrollably, at which point the crew were along for the ride, not having the necessary height to effect a recovery. Had they kept the speed at which they began the deacceleration they could have flown a circuit and landed. But that is hindsight, the crews performance was exemplary and unable to be faulted.

A work colleague from Pennsylvania, Dale Whitthoft, was a pax.

This was an accident that could not happen on a Boeing, as when slats are extended they are mechanically locked. But a Johannesburg 747 incident showed even their system had a previously unknown failure mode.
megan is offline   Reply With Quote
Old 29th Jan 2018, 13:09   #405 (permalink)
Thread Starter
 
Join Date: Nov 2015
Location: Seattle area
Posts: 208
I've uploaded a Youtube video of my animation of the crash at
.



There are two reasons for doing this. First, the flight sim run, and creating a virtual chunk of O'Hare, told me I'd drastically misunderstood what the plane must have done after the initial impact. I incorporated that sim run in this video. I put together the animation to help me understand the physics involved.



Second, the narrations I've given are an imprecise way to communicate what I envision happening.What I've posted isn't proof or evidence, just an artist's conception to help get the picture across.
BRDuBois is offline   Reply With Quote
Old 30th Jan 2018, 18:13   #406 (permalink)
Thread Starter
 
Join Date: Nov 2015
Location: Seattle area
Posts: 208
Interesting. Following this page link it's marked as removed by the user, but it plays fine. The link without the https is youtu.be/zfBQHyuPXGg Try that.

As a test, I'll create a fresh link on this post to the same video.

Hmm, seems happy. If you mouse over the yellow box in the preceding post where the video URL is shown, your nav status will show a different URL. Might be a board hiccup. So clicking the prior post's video was sending you off into never-never land.

Last edited by BRDuBois; 30th Jan 2018 at 19:46.
BRDuBois is offline   Reply With Quote
Old 30th Jan 2018, 20:06   #407 (permalink)
 
Join Date: Nov 2016
Location: Lakeside
Posts: 263
Hi.

In the air, even in extremis, aerodynamic surfaces do not lead, they follow. It is highly unlikely, IMO, that the remainder of the aircraft travelled at any time inverted and tail first. An intact tail will be feathers to the arrow, always. I continue to think the aircraft swapped ends and aspect late in the energy trail.

Your impact depiction is first rate.
Concours77 is offline   Reply With Quote
Old 30th Jan 2018, 20:34   #408 (permalink)
Thread Starter
 
Join Date: Nov 2015
Location: Seattle area
Posts: 208
I sure wouldn't have thought of it if I hadn't seen it in the sim. It's like a hydrofoil flipping over. Once the plane's longitudinal axis was vertical, it's all ballistics and not much in aerodynamics, I'd think. At that point everything is in stall, and the rotational momentum from the pitch-up continues until the tail is first.
BRDuBois is offline   Reply With Quote
Old 30th Jan 2018, 20:46   #409 (permalink)
Thread Starter
 
Join Date: Nov 2015
Location: Seattle area
Posts: 208
Here's a stunt plane going backwards.

BRDuBois is offline   Reply With Quote
Old 31st Jan 2018, 00:45   #410 (permalink)


Probationary PPRuNer
 
Join Date: Dec 2013
Location: Norfolk
Age: 61
Posts: 2
Excellent work with the potential break up sequences.

I have difficulty accepting that the tail section could have flown up and flipped vertically, landing upside down. Such a scenario would surely have crushed or badly damaged the vertical stabiliser. From the photographs, that appears the one piece of the aircraft that survived virtually intact.

Whatever mechanism inverted the tail section therefore had to be at a relatively low energy, just enough to flop over into the inverted position without destroying the vertical stabiliser.

This suggests a largely horizontal movement of pieces of the aircraft along or parallel to the ground, rather than cartwheeling high into the air.
G0ULI is offline   Reply With Quote
Old 31st Jan 2018, 01:42   #411 (permalink)
 
Join Date: Nov 2016
Location: Lakeside
Posts: 263
Two tail slides and a gentle Lomcevak. Flying tail first takes a lot of skill. Those maneuvers are all under complete control.

Even ballistic, drag and momentum have a part to play in trajectory. Your work is very skilled.
Concours77 is offline   Reply With Quote
Old 31st Jan 2018, 02:30   #412 (permalink)
Thread Starter
 
Join Date: Nov 2015
Location: Seattle area
Posts: 208
Quote:
Originally Posted by Concours77 View Post
An intact tail will be feathers to the arrow, always.
I understand your point, but you can throw an arrow backward like a dart and it will take eight or ten feet to turn around.
BRDuBois is offline   Reply With Quote
Old 31st Jan 2018, 18:49   #413 (permalink)
Thread Starter
 
Join Date: Nov 2015
Location: Seattle area
Posts: 208
I've uploaded a video showing how the sim behaves if the CG is moved five feet forward. It makes a drastic change, and the plane behaves much more as Concours77 was expecting. It is not how N137US behaved, of course.

The precise numbers are probably not important, since there's so much uncertainty in the sim in the first place. It's an approximation of the plane behavior, and I've drastically altered the plane to simulate the damage. So the sim's value is in showing broad trends and ideas.

As I say in the video, if this behavior is the explanation behind how N137US got to the final wreckage site, then N137US behaved more like a sim with the 20 foot CG displacement than 15, regardless that the real life displacement was some different value.

BRDuBois is offline   Reply With Quote
Old 5th Feb 2018, 19:57   #414 (permalink)
Thread Starter
 
Join Date: Nov 2015
Location: Seattle area
Posts: 208
I'm working on illustrations for my next published version, and just finished a couple examining the CG for the whole plane and for the plane after losing the engines and forward fuselage.

The whole plane graphic will be part of the knife edge flying discussion. I wanted to get a feel for the forces involved. It's at https://ibb.co/cZNvWH

The damaged plane graphic is to help understand the dynamics after the first impact as discussed above. I guesstimated that the CG would move aft about 20 feet. I guesstimated the center of pressure using a simple cross section area calc, which doesn't take into account things like drag around sharp edges. So I wouldn't trust those values within less than a few feet. But it shows that the new CG and CP are pretty much lined up, which is why the sim would be highly sensitive to moving the CG a small amount. It's at https://ibb.co/iWvSdx

The second one has a couple things I'll fix before putting it in the report.
BRDuBois is offline   Reply With Quote
Old 6th Feb 2018, 02:44   #415 (permalink)


Probationary PPRuNer
 
Join Date: Dec 2013
Location: Norfolk
Age: 61
Posts: 2
Craig

I fear you may be heading up a blind alley with these weight and balance figures. They are immensely important with regards to straight and level flight, but I have to question their relevance during the break up sequence.

Contact with the embankment removed several outboard feet of the starboard wing. Initial contact with level ground appears to have removed the nose and in all likelyhood most of the remainder of the starboard wing if the bank angle was actually around 90°, which of course you dispute.

This means that the remains of the aircraft would be very asymmetric, making a nonsense of any calculations based on all or part of an intact airframe.

It is necessary to start somewhere with the calculations, but I think this will be a much more complex matter than just considering a change in the centre of gravity caused by the loss of parts of the aircraft. I do not consider this can be a valid aproach to modelling the final disposition of the rear fuselage and wing box section unless it can be worked out for varying degrees of bank from level flight to at least 90° of bank. 10° increments would suffice to show the broad distribution of wreckage resulting and may contribute towards establishing exactly at what angle of bank the aircraft struck the ground. That is to say, whatever angle of bank calculation best fits the distribution of wreckage actually found at the scene will go some way to providing a definitive answer to whether the aircraft was levelling out or the bank remained completely uncontrollable.
G0ULI is offline   Reply With Quote
Old 6th Feb 2018, 18:56   #416 (permalink)
 
Join Date: Nov 2016
Location: Lakeside
Posts: 263
As to..

Agreed as to parameters of controlled flight. Motion of any body that has no aerodynamic stability, or stable CG cannot be predicted without a great deal of data. In the sixties, the supercomputer was an infant, and chaos was also in infancy.

The CAB had a purpose outlined in its remit. After the third impact, nothing was quantifiable such that a reward of any kind could be had relative to “preventing” such an accident in the future.

The single cable system was the underlying culprit, and nothing could be done to make it redundant, certainly not at reasonable cost. The “fix”, one assumes, is more frequent inspections (not terribly difficult) involving pulling up some carpet and removing an Inpanel. More stringent repair procedures were no doubt applied, etc.

No mention of the two terrible accidents that were seemingly cured by LEAP, and I find that strange. The experiment involving flap/aileron, done by Lockheed, in my opinion, was done to eliminate the possibility that the wing had been partially destroyed, either prior to, or as a result of, the last take off.

Howl as much as you like, Whirl Mode was on everyone’s mind, and purposely not bringing it up was a political move.

IMO
Concours77 is offline   Reply With Quote
Old 6th Feb 2018, 21:01   #417 (permalink)
Thread Starter
 
Join Date: Nov 2015
Location: Seattle area
Posts: 208
The graphic labeled "Knife edge flying" was intended to show the forces involved. It's interesting that it would take about 24,000 lbs of force on the rudder to do this, and I don't seriously contemplate that it happened or that the structure could stand it if it were attempted.

The balance issues for the broken plane are totally relevant. I agree that in a 90 degree bank more of the right wing would have been lost, but the CAB said it survived to WS 293, which is nearly to the number four nacelle. In the final wreckage the number three nacelle is clearly intact on its underside. The remains under where the rest of that wing burned confirms that there was wing past nacelle three. That's one of the best arguments that it wasn't in a vertical bank.

I have found varied evidence that the plane was in a shallow bank. If anyone wants to do the modelling to show what an Electra in a vertical bank would do, they're free to do so. The Argentine crash gives a good example of an Electra crashing in a vertical bank. My goal is to adduce evidence for what I think actually happened. I've run through many possibilities, now I'm working on presenting my case.

The purpose in running a sim on the broken plane was to find out what it would do in its damaged state. The result was so striking that I animated it that way, and it's very much like seeing a hydrofoil blow over. It's not a matter of treating the CG as the only important aspect, but of seeing what the CG change would do. It's certainly a valid question.

The difference between that and my animation of the CAB scenario is obvious, and I think anyone viewing the CAB animation would dismiss it. The sim is evidence (though thin) and the animation is illustration.

As for whirl mode, it was never a possibility. My mom got regular updates from people involved and she kept us informed. Whirl mode was not on the table in any way.

The sound of an engine in heavy whirl mode vibration is distinctive, the prop tips can go supersonic, and it was never reported. Everyone was aware that the Electra had a troubled history, and they wanted this solved fast so the public could be reassured. But whirl mode was not in the scope.
BRDuBois is offline   Reply With Quote
Old 7th Feb 2018, 00:10   #418 (permalink)
 
Join Date: Nov 2016
Location: Lakeside
Posts: 263
More...

Hello,

From the (sim) angles alone, my guess of angle of Bank at RR embankment impact is no more than forty degrees, likely less.

The angle of crossing included with angle of descent, plus dihedral makes sense only if the scar is in the vertical. A ninety degree bank is impossible, with the reported orientation of the airframe, as it involves flight characteristics that involve nonsense. If head on, the aircraft would have been ninety degrees PLUS dihedral.

That there was wing left out to the fourth nacelle shows the engine parted with the wing? The engine is not part of the wing structure, it rests atop the upper wing surface.

Had Whirl Mode been “on everybody’s mind”, CAB made a decision not to address it. You have evidence Whirl was actively eliminated, or just dismissed? “Well, we can eliminate Whirl Mode.....” Erm, exactly why?

I have proof via my own eyeballs that Whirl Mode continued through 1969. LEAP did not eliminate the phenomenon; it added structure that was supposed to prevent it from damaging the aircraft.

The gearbox/torque shaft/engine was a pile driver, capable of tearing the three connected structures free of the original mounts.

“We’ll add an additional torque strut, and embellish the mounts...” Lockheed

“The battery catches fire, ok. We’ll enclose it in a steel box....” Boeing
Concours77 is offline   Reply With Quote
Old 7th Feb 2018, 00:35   #419 (permalink)
 
Join Date: Mar 2005
Location: N/A
Posts: 1,966
There is a misunderstanding of whirl mode. Every engine has a whirl mode, it just requires the designing of adequate structure to prevent it becoming an issue. You can certainly see it at work in a 747 when in turbulence.

Whirl Mode: The Deadly Design Omission

http://ijme.us/cd_06/PDF/ENG%20204-005.pdf
megan is offline   Reply With Quote
Old 7th Feb 2018, 01:29   #420 (permalink)
 
Join Date: Nov 2016
Location: Lakeside
Posts: 263
Thanks megan, for the links. I am trying to suss my misunderstanding of Whirl Mode. I think I stated the phenomenon is common, and not unique to the Electra.

The concentration of power/mass locally (well forward of the wing) cannot be dismissed. The propellors are heavy, and have an unusually long moment arm related to the forward wing spar. It provides an extreme and asymmetric increase in power translated during whirl, relative to other installations. My experience with Whirl Mode occurred on short Final, as pilot selected full power to make up for an undershoot. The articulation of the nacelle was extreme, I promise you it was profoundly distressing to this passenger.
Concours77 is offline   Reply With Quote
Reply


Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off



All times are GMT. The time now is 23:49.


© 1996-2012 The Professional Pilots Rumour Network

SEO by vBSEO 3.6.1