Posted by: harrisonjones | June 26, 2014

NTSB report on Asiana at San Francisco

The NTSB has issued its initial findings on the Asiana accident at San Francisco. The recommendations are extensive, but not surprising. As expected, the probable cause is listed as the flight crew’s failure to maintain a stabilized approach, resulting in the aircraft being below the glide slope and below a safe airspeed. Contributing factors include improper training, crew fatigue, failure to adhere to established procedures, reliance on automation, and a host of other usual suspects.

Asiana

Let’s cut through all the jargon, acronyms, and modes to see if we can simplify this report. The B-777 has two throttles on the control pedestal, and as you may correctly surmise, it has two engines. Push throttles forward and airplane goes fast. Pull throttles aft, airplane goes slow. Yeah, I know, pitch attitude is involved and blah, blah, blah, but let’s keep it simple. There are two entities that can push the throttles forward when the airspeed is low. One is the pilot and the other is the auto throttles. Someone has to take charge and push. The pilot can override the auto system at any time by simply pushing the throttles, no matter what mode the system is in. The NTSB says the pilot should have done just that to avoid the catastrophe. Draw your own conclusions.

There are interesting side issues revealed in the findings:

Two emergency evacuation slide/rafts (1right and 2right) inflated inside the airplane on impact. I can see how that would happen due to the impact force. If the door moves with the slide armed, the slide will inflate whether the door actually opens or not. The door at 4right actually separated from the airplane and struck a seated passenger, causing severe injury.

Two passengers were ejected from the cabin, when the tail separated from the fuselage because they were not wearing seat belts. Don’t know what to say about that. Seat belt signs, PA announcements, safety videos, etc. It’s an excellent idea to wear a seat belt.

Four flight attendants, seated in the aft galley area, were ejected when the tail separated even though they were strapped in. When the tail is ripped off, things happen. They still survived. It’s an excellent idea to wear a seat belt. Pardon my redundancy.

When the airplane came to a stop, the captain made an announcement and ordered everyone to remain seated. Obviously those who were ejected cannot be held accountable for premature evacuation. After 90 seconds, a flight attendant saw fire and smoke and initiated the evacuation anyway…rightly so. A minor point to notice here is that the emergency electrical bus, powered by the battery, remained operational and enabled the PA system. Radio communication with the tower remained operational also. Boeing builds a fine airplane.

One of the ejected passengers was subsequently run over by a fire truck, not once but twice. The NTSB recommends training for firefighters to watch out for passengers who have evacuated. Probably a good idea, but this is not as simple as it seems. Fire, smoke, fog, night conditions, fire fighting foam, adrenaline-charged firemen, panicked passengers, etc. complicate the issue. In this case, firemen entered the burning aircraft and removed five injured passengers who could not evacuate without assistance. God bless them.

According to the NTSB, the crew became aware of the low and slow condition at 200 feet above the ground, but did not initiate a go around until below 100 feet. At a normal descent rate of 700 feet per minute, that’s a delay of about 10 seconds. They did mention that fatigue was a factor, but even so…As a measuring stick; I might mention that go arounds are routinely initiated from 50 feet in the simulator (a truck appears crossing the runway) however that assumes a normal approach speed and rate of descent.

I’ll leave the CRM discussion to those enthusiasts who would have formed a committee at 200 feet and argued the merits of a missed approach.

I admire the NTSB for their expertise and thoroughness, and I applaud Boeing for designing and manufacturing an airplane that can sustain the force and damage of this accident and still result in a 98% survivability rate. The complicated discussion of fatigue, automation complacency, and CRM will continue. The simple laws of thrust, drag, lift and weight will remain constant.


Leave a comment

Categories