American Morning

Tune in at 6am Eastern for all the news you need to start your day.
May 13th, 2009
11:12 AM ET

Mandatory training needed to prevent pilot error

[cnn-photo-caption image= caption= "Greg Feith is a former senior investigator for the National Transportation Safety Board."]

Lawmakers on Capitol Hill today are investigating the deadly crash of Continental Connection Flight 3407, which killed 50 people in February near Buffalo, New York.

In a story Monday, the Wall Street Journal cited investigators as saying the crash resulted from pilot Marvin Renslow's incorrect response to the plane's precarious drop in speed: He overrode an emergency system known as a "stick pusher," which sends the plane into a dive so it can regain speed and avoid a stall.

Colgan Air, the operator of Continental Connection flights, said Monday that Renslow had never trained in a flight simulator with the safety system that activated just before the plane went down. Colgan said there is no regulatory requirement that it provide hands-on training with the "stick pusher."

"A stick pusher demonstrated in an aircraft simulator is not required by the FAA," the airline said in a statement. "And thus was not included in Colgan's Q400 training program."

The Federal Aviation Administration said its standards do not require hands-on practice with the safety system.

Greg Feith, a former senior investigator for the National Transportation Safety Board, says flight simulation training on this type of system should be required for all pilots. He spoke to Kiran Chetry on CNN’s “American Morning” Wednesday.

Kiran Chetry: The cockpit voice recorder revealed a conversation took place five minutes before the crash. First Officer Rebecca Shaw said she had “never seen icing conditions.” She said “I don't want to have to experience that and make those kinds of calls. You know I’d have freaked out… I'd have seen this much ice and thought, ‘Oh my gosh, we were going to crash.’” She's telling the pilot she's worried about icing conditions. It seems like a scary conversation to be having in the cockpit at the time.

Greg Feith: Absolutely. When you look at the transcript and where that conversation took place being so close to the final event, you have to wonder why their awareness wasn't higher when they first took off and got in to the icing conditions. And one of the things that the NTSB is going to really have to look at is why they breached that sterile cockpit rule. But if you look at the transcript, they talk about ice, and then they go back to their normal conversation. They don't really talk about the flying of the airplane and the approach speeds that they need to be flying.

Chetry: When you say sterile cockpit, it's the FAA rule that forbids any non-flight related talk from happening below 10,000 feet. But we’re hearing casual conversation. People talk about being tired, maybe they should have called out sick. How common is it to hear that casual conversation on an approach?

Feith: Well, we don't really hear it because unless the airplane has an event that the NTSB or the FAA has to investigate, we don't know. But I've ridden in the cockpits, and there is casual conversation that takes place. But it's the discipline here. I mean, you’re in a critical icing situation, you're going to be shooting an approach down to what we call minimums, the weather is very bad. You don't want to be talking about anything other than flight-related duties. And that is the big issue here is they didn't even think about that when - after they talked about ice. They went back to the normal conversation that was non-flight related.

Chetry: Some are saying the critical mistake made was that the captain never was properly trained on the plane's anti-stall “stick pusher.” This is a safety feature that automatically points the plane's nose in to a dive so it gains speed to prevent a stall if it slows down. Pilots are apparently trained they're supposed to push the stick forward to gain speed if it happens. But he apparently yanked back on it at the time, causing the crash. Was this preventable in your opinion?

Feith: Well, when you look at it, Kiran, you have to look at the training program. The training programs in a lot of the airlines don't want you - don't want a pilot ever to get into that stalled condition. So they train you to recognize when the airplane is getting slow and getting in to that stalled condition before it actually gets there. So they want you to recover and be away from it. So, for this pilot not to have gotten the full training of this stick pusher… But the bigger thing here is because he probably knew he was very low to the ground, when the nose went over, his immediate reaction not necessarily to the stall and the proper technique to recover from it, was probably more of an emotional reaction that he knew he was close to the ground and it was an instinctual move to pull the nose back up to level so that he wouldn't hit the ground.

Chetry: Was it the wrong move in your opinion?

Feith: Well right now it's real easy to sit back and say it was the wrong move. But I can understand why he probably did that. He didn't have a full comprehension of what was going on with the airplane. He knew he was close to the ground. All he was trying to do was get it back to level flight.

Chetry: Apparently it's not mandatory at this point to be trained on that in terms of flight simulators versus what you just learn in the books on the ground. Should this be mandatory?

Feith: Absolutely. And I think that pilots, so that they can build the mental modeling and the tactile skill development should be taken all the way through full blown stalls in the simulator because the fidelity of the simulator is pretty good. But they should at least be exposed to it so that they have a total awareness. Don't just take them up and expect them to read about the fact that this is what's going to happen if you do get into a full blown stall.

Filed under: Transportation
soundoff (2 Responses)
  1. JP

    The bigger problem lies with inexperienced crews of commuter airlines. Due to recent shortage of pilots, commuter airlines were/are hiring pilots with only few hundred hours of flying time. However the biggest problem is that most of these pilots have gained experience only in flight instructing and have no practical pilot in command hands on experience. Instructor sitting on the right seat of a trainer can log all this time as pilot in command when in fact he/she might not even handle the controls. This same instructor might never leave the airport traffic pattern/practice area and keeps doing same type of flying for years to build “flying time”. This makes him/her an experienced flight instructor but a lousy pilot. As an example flight instructor might have gained enough flying hours to qualify as a captain in commuter operations but is completely lacking knowledge of actual flying in hazardous all weather conditions and decision making. Due to lack of experienced pilots commuters upgrade pilots to captains with only FAA mandated minimum hours and require passing of upgrade training and check ride. This is painfully clear in this accident that should have never happened. The captain should have never been allowed to be a captain in this complex aircraft.
    There has to be a change how instructors log flying time as pilot in command. Instructor time and actual pilot in command time have to be separated. Current flying time logged by instructors should be prorated versus actual flying time as pilot in command. There has to be higher total time requirements to qualify as a captain in commuter operations.

    May 13, 2009 at 1:31 pm |
  2. Jackie in Dallas

    It is always easier to have 20/20 vision in hindsight, as Feith says. What is not easier is to make assumptions based on recorded conversations. Have you never made nervous conversations when faced with a stressful situation? I know I have. You put me into an airplane cockpit where the plane looks to be in trouble, and you bet I might be saying "I should have called in sick today!"

    The key issue is pilot training. While it is not always feasible to train pilots on every possible scenario (namely because some will come up that noone has ever seen before), it does appear that the training was faulty. But Feith is also correct, probably, in assuming that the pilot knew how close to the ground he was, and when the automatic system sent the plane into a dive, he instinctively responded.

    The other real issue was whether that automatic system was including the distance to the ground in the computerized response. THAT is the question they should be asking, as well. What's the difference between crashing from a stall, and crashing by sending the plane nose down into the ground? The result is the same...dead passengers and crew. That is the story I want to hear - did the software designers and pilots working with them take THAT into account in the automated response?

    May 13, 2009 at 12:53 pm |