Plane Down in Buffalo - Colgan Continental Flight 3407

[Velocipede]

Just caught up on this thread and I have a few thoughts for all of you, so here comes one of my famous "multi-quotes".

I think there's a LOT to learn from accidents that happen, and for us to study the transcripts & the readings from the FDRs is a great learning experience. But on a human level, I do not find any value in the sensationalism of hearing the actual recordings of people's "last words".

You are absolutely correct. Everyone can learn from CVR transcripts, but only ghouls and TV reporters want to actually hear the tapes. I've listened to some CVR tapes and its very distressing. No one really gains anything more from listening to the tape recordings EXCEPT to determine if something was forgotten.

Example: In the AA crash in Little Rock, the CVR tapes were critical in determining that the Captain forgot to arm the ground spoilers. In the DC-9/MD-80 the spoiler handle makes a very distinctive click noise when its armed. That was missing from the tapes and not having automatic deployment of those spoilers was a heavy contributing cause to that airplane going off the end of the runway.

But the briefing does raise questions. At some point the nose had to be quite low to descend so rapidly or the earlier briefing about 'severe pitch and roll' changes makes no sense.

I have to agree with Cruise here. We're starting to see some speculation as to cause on this thread. Let me remind you we do NOT have the evidence to make ANY judgments about what happened. All I can tell you guys is I've been involved in accident investigation since my Navy days and even professional investigators usually have to go through several scenarios before they figure out what really happened. And THEY have the evidence in hand. We don't.

I'd gently suggest that there are enough people who read this board who have been emotionally hammered by this incident that we really don't need to speculate. Let's let the NTSB do their job. Once the evidence is in, we can debate it. Until then, idle speculation makes us no better than those ******* trolls on TV.

If these 4.2 hour student pilots are expert enough to know the fundamental aerodynamics of airplanes and icing then how can 2000+ hour commercial pilots not be expert enough to fly them?

Listen, pal. Every single time you fly in the weather, the environmental conditions are different. And I'd venture to say there are 18,000 hour pilots who are not "expert" aerodynamicists. So let's leave those kind of comments where they belong...in the trash can.

I think you're BOTH speculating. And, quite possibly both WRONG.

Edit: My apologies if this came out harsh.....been a long couple of days w/ many more to come.

:yeahthat:

And no need for apologies. You are at the point of the spear on this one.
BTDT (AS261) and I agree with you completely.

I'm not really speculating...just making some comments.

Generally, this is the opening comment to rampant speculation.

If the accident occurred after the flaps went to 15...that does not seem like a large enough flap setting to trigger a tail stall. If that is what occurred, it'll be surprising to me that flaps 15 induced it.

Go back and read my description of my tail stall pitchover in the Convair. Flaps 15 was exactly what precipitated that event. So, this speculation of yours is incorrect.

I think it is also a possibility that the flaps extended asymmetrically.

Speculation. What evidence can you present to support this comment. None, because you haven't actually seen any of the wreckage.

Also, this accident seems very similar to the CMR crash in Detroit that went into an accelerated stall while joining the localizer during icing conditions.

Again, speculation. This time linking a different airframe to this incident. All in all, this whole post was speculation in spite of your intial "disclaimer."

The night before, I reported moderate icing to ATC while flying out of ILN, changed my mind, and told them to change it to severe in my PIREP..

Just a question. Is it legal for you to fly in severe icing? Is your airplane certified for it?

Of course, experience level does factor into the question of whether a pilot could've prevented an accident--either by act of commission or omission. Thus, experience level can be a contributing factor in any accident.

Entirely correct.

Do I think this difference will be highlighted in the inevitable wrongful death trials? Absolutely. I guarantee some trial lawyer will ask a jury to ponder this question: "Continental Airlines is selling the experience level of a Continental Airlines crew to its customers, yet delivering that of a Colgan crew without the customer's knowledge or consent; therefore, isn't the risk imposed by this substitution entirely Continental's?"

Interesting thought. However, I'm wondering if you could even make Continental a party to the suit. They weren't operating the airplane. They didn't manufacture the airplane. I guess a smart lawyer could make that argument, however just in the case of the accident investigation itself, Continental would not even be admitted as an "interested" party. That would be limited to ALPA (pilots), Colgan (operator), and Bombardier (manufacturer) in addition to FAA (ATC) and the NTSB.

Ignoring "eyewitness accounts" from the ground seems to be a good way to find out what actually happened in an aviation crash investigation either because the "eyewitness" knows very little about airplanes, the way they fly, and what causes them to not to, or because the "accounts" never were "eyewitnessed" in the first place, rather a cheap way to get on tv.

Possibly. But as an investigator, you interview EVERYBODY looking for any information that could help determine the cause. You have to seperate a lot of chaff to get very little wheat, but sometimes its worth it. An interesting aside...you'd be surprised how many people are taking pictures of airplanes. In this case it was nighttime and poor weather, but one of the thing investigators are always looking for is someone who may have snapped a picture of an incident aircraft.

My Grandpa always said that in the training center at least, the younger low-time kids were easier to train for the jets than the guys who had cut their teeth on DC-3's, Martin's, etc - fewer ingrained habits to break.

Some of the toughest Naval Aviators to train on the DC-9 and CV-580 were the fighter/attack pilots. They were so used to doing everything in a single pilot environment that some of them had trouble switching over to a crew served airplane. That mindset from years of training and successful operation was sometimes hard to redirect.

The YS-11 would begin 'moaning' when it was picking up ice and shortly after you heard the 'moan' you would get the ice being thrown into the fuselage and off the props. Disconcerting the first time you heard it and attention getting from then on.

Believe it or not, we had a maintenance profile in the DC-9 where you actually had to go out and stall the airplane (not a fun event, btw). But the -9 would do the same thing the last 10 kts before it stalled...moaned. It was an eerie sound. One you didn't want to hear during normal ops because when it stopped something BAD happened.

 

[beechpilot]

I can't recall an eyewitness account of any crash that didn't involve "the engine was sputtering" or "it was on fire before it hit the ground", when almost always engine trouble and/or fire damage had absolutely nothing to do with it.

Ignoring "eyewitness accounts" from the ground seems to be a good way to find out what actually happened in an aviation crash investigation either because the "eyewitness" knows very little about airplanes, the way they fly, and what causes them to not to, or because the "accounts" never were "eyewitnessed" in the first place, rather a cheap way to get on tv.

$.02

I agree that many witnesses won't have much to offer but this statement isn't always true. Last year I was working when I watched a series of events unfold just prior to a fatal crash of a Citation. I was involved heavily in the investigation and what I was able to tell investigators helped them paint a picture of what happened. Had I turned my back and not paid attention there would have been little to no information as to why the accident may of occured.

 

[WacoFan]

Believe it or not, we had a maintenance profile in the DC-9 where you actually had to go out and stall the airplane (not a fun event, btw). But the -9 would do the same thing the last 10 kts before it stalled...moaned. It was an eerie sound. One you didn't want to hear during normal ops because when it stopped something BAD happened.

My Grandpa loved the DC-9. He was kind of TWA's point person on it when they bought them - wrote the manual, developed the training, and did the acceptance flights. I guess that was one of the early airplanes that had a split elevator and he flew an early one that was having a problem that would cause an asymmetric condition in the elevators. Anyway, he, a copilot and the Chief pilot flew it on a maintenance flight and while slowing to do a stall somehow found themselves upside down after a half snap-roll. Grandpa did a slow roll to recover and everything everything worked out ok. Only heard him tell the story once and it was LONG ago but it was really interesting to hear him talk about the transition from props to jets and the training challenges that presented. Also, given the crashes in the 60's that took place, many that were pilot error, it gives you a new appreciation for the safety records that are accomplished today as a result of technology and really excellent training that has been developed over the decades from hard experience.

 

[Velocipede]

Waco,

Thats how a -9 stalls. You get the SSRS and then one wing (and there's no predicting which one!) drops, followed by the nose. Takes about 6000' to recover. We always did them 15k agl just for that reason.

I think I did two of those test profiles. After that I decided you could probably test the SSRS just as efficiently in the hangar with a broom handle on the stall vanes. I put in a change to the NATOPS manual to do just that. It took a while, but they changed the test profile and we didn't have to stall the airplanes anymore.

It was scary, even if you knew what was going to happen.

 

[MikeD]

I didn't see anyone questioning the crews experience either, but I think things were getting side tracked with another "experience" debate, which I think is only respectful that we keep that kind of debate for another thread. Most people are coming to this thread to discuss their feelings and get things off their chest. It would be nice if it stayed that way.

Which is why I kept it on track, instead of debating the rights or wrongs of experience; instead discussing how it may affect things in this accident...as it relates to the accident, not the crew in question. It's the natural progression of a thread, covering all the varying aspects and potentials of an incident of this type. Done respectfully, it can integrate with the rest of the discussion on this subject. If it becomes it's own singular debate point or a flame war of any kind, then I think that'd be the time to send it to it's own thread or otherwise have it be dealt with.

 

[SlumTodd_Millionaire]

Waco,

Thats how a -9 stalls. You get the SSRS and then one wing (and there's no predicting which one!) drops, followed by the nose. Takes about 6000' to recover. We always did them 15k agl just for that reason.

I think I did two of those test profiles. After that I decided you could probably test the SSRS just as efficiently in the hangar with a broom handle on the stall vanes. I put in a change to the NATOPS manual to do just that. It took a while, but they changed the test profile and we didn't have to stall the airplanes anymore.

It was scary, even if you knew what was going to happen.

Holy crap, that's a scary way to test an airplane. I've certainly never stalled the 717 (same airframe as a DC-9-30), but knowing how it handles just under normal conditions, I definitely wouldn't want to try stalling it. Damn.

 

[Snoopy2]

Ignoring "eyewitness accounts" from the ground seems to be a good way to find out what actually happened in an aviation crash investigation either because the "eyewitness" knows very little about airplanes, the way they fly, and what causes them to not to, or because the "accounts" never were "eyewitnessed" in the first place, rather a cheap way to get on tv.

No one is claiming that eyewitness accounts are evidence of anything. However this eyewitness actually saw the plane and was quite accurate in describing the direction 3407 was flying right before impact, i.e. away from the airport.
The NTSB account confirms his testimony.
Not every eyewitness account is wrong and not every opinion offered up by a member of the aviation community is right.

 

[kellwolf]

The next time Im in icing will be much different too. Autopilot is coming off and will be hand flown, and I feel like Ill react faster if something does happen.

I thought about doing this in the CRJ for about 20 seconds. Then the systems knowledge kicked in. With all the hydraulics and dampening on the CRJ, you probably wouldn't feel any different forces in severe icing as you would in normal flight. In fact, there are cases of the aircraft stalling BEFORE getting a shaker/pusher b/c of how the stall system works. With the ice on the wing, the wing stalls at a different angle of attack. The stall system works off the AOA vanes on the side of the airplane. If they don't have the exact same amount of icing on them as the wings, you won't get any warning. That's the reason we don't go to 15 degrees of pitch up on t/o anymore. They had some people rotating with ice on the wings, and the aircraft stalled without giving them any warning.

The Saab is probably different since it probably still has cables connected to the control surfaces. In that instance, you'd probably feel a different control force in icing conditions.

 

[WacoFan]

Holy crap, that's a scary way to test an airplane. I've certainly never stalled the 717 (same airframe as a DC-9-30), but knowing how it handles just under normal conditions, I definitely wouldn't want to try stalling it. Damn.

You should have heard him talk about something in the Connie called "the busy drill". It was a simulated two-engine failure on take-off while dirty. Had to get the airplane cleaned up, fly the approach and land. I forget the details but have a written description somewhere and will try to dig it up. Of course, all training was done in the airplane then so that was neat too. If I can find that description I'll send it.

 

[kellwolf]

Holy crap, that's a scary way to test an airplane. I've certainly never stalled the 717 (same airframe as a DC-9-30), but knowing how it handles just under normal conditions, I definitely wouldn't want to try stalling it. Damn.

They want anyone here to actually stall an airplane, they can call Gison. I ain't doin' it. They even just took the stall profiles out of the CFM. Not sure if that's a good or a bad thing.

 

[Rocketman99]

I would see that as a bad thing. I can totally understand not doing them in the actual aircraft, it just makes sense. However, not to do them at all, even in the sim? That sounds like just asking for trouble. I've flown several large of the jet and prop category and none of them quite the same. The worst I can think of was the pre-stall buffet in the KC-135. Everything in the console just starts creaking and shaking violently along with the plane as it approaches a stall. You couldn't miss it. I'm sure we all know the basic premises of the types of stalls, but knowing the specific characteristics of the airplane you're flying is beneficial IMO.

 

[AMH]

Possibly. But as an investigator, you interview EVERYBODY looking for any information that could help determine the cause. You have to seperate a lot of chaff to get very little wheat, but sometimes its worth it. An interesting aside...you'd be surprised how many people are taking pictures of airplanes. In this case it was nighttime and poor weather, but one of the thing investigators are always looking for is someone who may have snapped a picture of an incident aircraft.

A picture is worth a thousand words and cameras don't lie.

No one is claiming that eyewitness accounts are evidence of anything. However this eyewitness actually saw the plane and was quite accurate in describing the direction 3407 was flying right before impact, i.e. away from the airport.
The NTSB account confirms his testimony.
Not every eyewitness account is wrong and not every opinion offered up by a member of the aviation community is right.

I don't doubt that at all. The aircraft was facing northeast on the ground, while the runway is facing southwest. Seems like a no brainer. "Sputtering" on the other hand seems inaccurate. Although, the full report is yet to come, the NTSB has stated that the engines seemed to be in working order. The NTSB is usually a more reliable source than some dude.

 

[SeanD]

Breaking news on MSNBC. The airplane was on autopilot.

http://www.msnbc.msn.com/id/29194109/

 

[Velocipede]

Holy crap, that's a scary way to test an airplane. I've certainly never stalled the 717 (same airframe as a DC-9-30), but knowing how it handles just under normal conditions, I definitely wouldn't want to try stalling it. Damn.

Yeah, after the second time I had to do it, I thought..."This is stupid" and decided to put in the NATOPS change. Apparently, the powers that be agreed with me.

I don't doubt that at all. The aircraft was facing northeast on the ground, while the runway is facing southwest. Seems like a no brainer.

There's lots of things that could get you facing the wrong way...a spin, for instance. And from 1500-2000 feet in an airplane that size, you'd get about half a turn. That would point you away from the airport.

And spins are funny things. You don't go straight nose down until the spin has been fully developed. My spin training was in the T-28 and the nose would oscillate up and down until the spin was fully developed.

Which leads to the question...How many civilian trained pilots get spin training? In the Navy, we had two spin hops in basic and I believe the jet guys had even more spin hops in advanced flight school.

Spinning is actually kind of fun IF you start high enough to recover. I accidently put myself in an inverted spin in the T-28 once. Tried to do an Immelmann except I started the loop part about 20 kts too slow. Spun out at the top inverted. THAT was a trip, man.

Of course, I scared myself and decided to call it quits for the day. The Mainentance Chief said it was the shortest student solo (about .8 hour as I recall) that he'd ever seen logged.

 

[B767Driver]

Again, speculation. This time linking a different airframe to this incident. All in all, this whole post was speculation in spite of your intial "disclaimer."

Velo...thank you so much for getting out your red pen and setting me straight. Let me reword my submission and see if it meets your guidelines for submission.

This accident has caused me to reflect on some other loss of control accidents that have previously occurred and I feel are relevant to consider in everyday flying. In regards to flying in icing conditions, a tail plane stall is definitely something to consider. Flap extension increases downwash and consequently the AOA on the tail. I would be more cautious adding flaps beyond approach flaps...but any amount of flap extension should be met with caution. I would be interested to know how much the downwash angle is affected by selecting approach flaps or less.

Also while flying in icing conditions, an increased speed should be considered and bank angle limits monitored...as the risk of an accelerated stall during maneuvers such as a course intercept are completed.

As I recall, the FAA issued an AD for flying turboprops in icing conditions about 12 years ago that limited flap settings and imposed minimum speeds while operating in icing conditions.

Also, anytime flaps are selected, I think it's a good technique to monitor the flap gage for symmetrical extension. Asymmetrical extension can cause a rapid, possibly unrecoverable, roll onset.

 

[staplegun]

Waco,

Thats how a -9 stalls. You get the SSRS and then one wing (and there's no predicting which one!) drops, followed by the nose. Takes about 6000' to recover. We always did them 15k agl just for that reason.

I think I did two of those test profiles. After that I decided you could probably test the SSRS just as efficiently in the hangar with a broom handle on the stall vanes. I put in a change to the NATOPS manual to do just that. It took a while, but they changed the test profile and we didn't have to stall the airplanes anymore.

It was scary, even if you knew what was going to happen.

Velo-

I flew C-9B's and did that stall test many times. I agree it was scary, but it was also under "controlled" circumstances - you knew what was coming.

I learned a lot about flying transport-category aircraft form my C-9 experience. Stalling, single-engine, touch-&-go's - all performed in the actual airplane.

There is a whole generation of pilots who have never experienced these things other than in the simulator. The sim is a great training/teaching tool, but there is no comparison to actually doing it in the aircraft.

I knew one of the Airborne pilots who was killed while doing an acceptance check flight on a DC-8. He was a simulator instructor on their DC-8. The crew stalled the airplane as part of their acceptance procedure - unfortunately their DC-8 simulator was not programmed correctly and so the procedures they used in the simulator to recover from a stall were not quite the same as what it actually took in the airplane, so 6 people died.

I think that every pilot who is serious about being an aviator should, at some time in their career, take an airplane to the limits of its flight envelope. I think it instills confidence in the pilot that 1) the aircraft will perform as advertised and 2) the pilot can control the aircraft even at the edge of the envelope.


JMO...

Kevin

 

[beechpilot]

Which leads to the question...How many civilian trained pilots get spin training?

If anyone goes for a CFI they have to complete spin training. I did 24 spins for my training and learned a lot from doing it.

 

[Orange Anchor]

I have to agree with Cruise here. We're starting to see some speculation as to cause on this thread. Let me remind you we do NOT have the evidence to make ANY judgments about what happened. All I can tell you guys is I've been involved in accident investigation since my Navy days and even professional investigators usually have to go through several scenarios before they figure out what really happened. And THEY have the evidence in hand. We don't.

So, I take it that you would agree with Dekker's observations that cause is not so much found as constructed?

Believe it or not, we had a maintenance profile in the DC-9 where you actually had to go out and stall the airplane (not a fun event, btw). But the -9 would do the same thing the last 10 kts before it stalled...moaned. It was an eerie sound. One you didn't want to hear during normal ops because when it stopped something BAD happened.

When Southern (yes a long time ago) got their DC-9 they were still doing flt training in the airplane and reportedly took one out and started the stall series. They deep stalled the airplane and almost did not recover. I am surprised you guys stall a T-tail. I've done full aft stick stalls in the Lears with the ventrals and no big deal but straight T-tail? No thank you.

 

[cj4pilot]

In most recently built Citations, Stick Shaker Activation goes up about 10-15 kts(activated at lower AOA, happens sooner) with Airframe anti-ice selected on. Also you can research a Citation 560 crash in Colorado. Icing conditions present. 2 Circuit City Citation 560's getting vectored for the approach, one right behind each other. If I am not mistaken the first one flew the approach 7-10 kts slower than the second. The first one didn't make it. Because of 7 kts!!!!! As a pilot and mechanic, I keep in mind whenever I am in Icing conditions I am a Test Pilot. My paycheck does not say Test Pilot so I do my best to retreat toward conditions benefiting my pay grade. No ice buildup is ever exactly the same, so your airfoil is never exactly the way the FAA/Manufacturer tested it. Not to mention the other variable, weight. One great help is that I have never had to abide by an airline's schedule, just a boss who understands that the utility of his aircraft is not limitless.

Disclaimer

---Ice may or may not have affected flight 3407 a DHC-8-400Q an aircraft which I have no intimate knowledge of handling or systems. The holes of the swiss cheese have to line up some how. Most aviation professionals know that there can be a large number of casual factors(MX/System Failures etc.)leading up to an accident or incident, and judgment before the experts(NTSB not media) weigh in is pointless. ---

 

[Orange Anchor]

The worst I can think of was the pre-stall buffet in the KC-135. Everything in the console just starts creaking and shaking violently along with the plane as it approaches a stall. You couldn't miss it. I'm sure we all know the basic premises of the types of stalls, but knowing the specific characteristics of the airplane you're flying is beneficial IMO.

I've done full aft stick stalls in the 737-300 and it is the same. First an aero-buzz which gets louder and then the airplane begins buffeting and finally the nose falls through like a big Cessna (if there is such a thing ) . The airplane will tell you it is not happy and is not going to continue to fly if you persist.

But it is obviously not a T-tail.