The Investigation and Findings in the Crash of US Air Flight 405
The aircraft accident I am reporting on is USAir flight 405. The crash of USAir flight 405 is like most aircraft accidents in that there were several contributing factors that led up to the accident. This report deals with procedures and pilot error that were deemed by the NTSB to be inadequate and/or improper. The adverse weather conditions, deicing procedures, take off procedures and take off delays were all contributing factors that led up to the take off stall in icing conditions that resulted in the fatal crash of flight 405. The captain, one cabin crew and 25 passengers received fatal injuries. There were 22 passengers, one cabin crew and the first officer that survived the accident. The survival of the first officer, cabin crew and passengers provided important first hand accident data.
On Sunday March 22, 1992 about 9:35 eastern standard time, a Fokker 28-4000 series, tail number 485 uniform sierra, crashed during an attempted take off from runway one three at La Guardia Airport in Flushing, New York. The weather observations at La Guardia during the time of the accident were instrument meteorological conditions which included 700 foot obscured ceilings, visibility �¾ mile, light snow and fog, wind 070 at 13 knots, temperature 31 degrees F., dewpoint30 degrees F., altimeter 29.67, runway visual range 6,000 feet plus, drifting snow, wet snow, snow increasing one inch in the past hour, and one inch on the ground.
The aircraft crashed on the attempted take off due to ice contamination on the wings which resulted in an aerodynamic stall and loss of control after liftoff. The aircraft came to rest in the waters of Flushing Bay partially inverted with parts of the aircraft fuselage and cockpit partially submerged in the water. There were several passengers that survived the initial impact but ultimately expired due to smoke inhalation and some drowned in the Bay.
Us air flight 405 was late arriving at La Guardia from Jacksonville, Florida because of weather delays and was preparing for the next leg of the trip to Cleveland, Ohio. The aircraft was parked at gate 1 and was cleared by the line mechanic as “good to go.” The captain and first officer then left the aircraft and went into the terminal for about 15 minutes and then returned to prepare for departure. Neither the captain nor the first officer performed a walk around preflight inspection, nor were they required to do so by USAir procedures.
Prior to pushback the aircraft was deiced with USAir deicing equipment. Two trucks using deicing fluid type I with a 50/50 glycol/water mixture was used to deice the aircraft at 8:26 pm. One of these two trucks experienced mechanical problems and blocked the aircraft which caused a 20 minute push back delay. The captain ordered another deicing by the single truck and was completed by 9:00 pm. At 9:05pm the first officer contacted ground for taxi clearance and changed to the tower frequency at 9:25. The captain announced that they would be using USAir’s contaminated runway procedures which called for 18 degrees of flaps.
The captain then stated that they would use a reduced V1 speed of 110 knots. The first officer completed the before take off checklist during taxi. The first officer later said that he had used the ice inspection light at least three times prior to take off. He used the inspection light to check the black strip on the leading edge of the wing and said that he had not seen any ice contamination. The NTSB investigators and other parties met at Newark International Airport, Newark, New Jersey, to observe the F-28 wings at night before and after deicing. They observed the wing of the F-28 with the same inspection light with the first officer’s window open and closed.
They found that the inspection light was little help in detecting ice on the upper portion of the wing and it was even more difficult to detect ice with the window closed. The first officer recalled that he had checked the wings with the windows closed. An example of improved aircraft technologies would have been a second black strip on the upper surface of the wing to detect ice on the ground because the existing black strip on the leading edge was only for detecting in flight ice. The cockpit voice recorder showed that flight 405 was cleared into position and hold at 9:34 on runway one three. Then the airplane was cleared for take off around 9:35. The take off was initiated only seconds later and the first officer recalled that the take off was normal through rotation but never recalled a positive rate of climb.
The cockpit voice recorder recorded the first officer calling out the V1 speed and seconds later recorded the Vr rotation callout. At 9:35:28 the cockpit voice recorder recorded the sound of the nose strut extension about 2.2 seconds after the rotation callout and then 4.8 seconds later the Cockpit voice recorder recorded the stick shaker which continued until the end of the recording. At 9:35:33 the first stall warning sounded and then 4.9 seconds later a series of five stall warnings went off. At 9:35:42 the sound of initial impact was recorded and then the recording ended. The first officer said that as they were flying in ground effect a pronounced buffet developed and the airplane began to roll left. The first officer stated that he joined the captain in leveling the wings and that they both knew that the aircraft would not fly and headed towards the blackness of the water. The aircraft broke into four primary sections and came to rest partially inverted at the edge of Flushing Bay and several fires broke out.
The initial delay and taxi time of 35 minutes exceeded the prescribed 11 minute hold over time for type I deicing fluids. During that 35 minute time period the airplane had accumulated ice contamination on its lifting surfaces. The FAA mounted what they called a “sharply focused effort” and convened on May 28 and 29 1992 in Reston, Virginia for the International Conference on Ground Deicing. There were leading experts from around the world which discussed industry methods and actions to be taken in the long and short term. They discussed different types of deicing fluids which were not used at the time of the accident in La Guardia and they also discussed different deicing equipment and techniques. They also found that the pilot in command was the ultimate authority for take off decisions but that all part 121 operators had to provide proper criteria to follow in order for the pilot in command to base a proper decision.
The new FAA rule required that all part 121 operators put in place FAA approved ground deicing or anti-icing procedures anytime weather conditions of ice, snow or frost prevailed. The new rules went into effect on November 1st 1992. These procedures, regulations and industry efforts that came from this conference are similar to other safety initiatives that occur only in response to an accident instead of a pre-emptive measure.
At the request of the NTSB Fokker conducted a special study of the effects of ice contamination and pilot technique on the F-28 aircraft. The NTSB evaluated the data from the Fokker wind tunnel and came to the conclusion that the captain rotated the aircraft five knots early at 119 knots instead of the proper rotation speed of 124 knots. The wind tunnel data also concluded that particles as small as one millimeter to two millimeters can cause a loss of lift of 22 percent in ground effect and 33 percent out of ground effect. The data from Fokker was correlated with the cockpit voice recording and confirmed that the first officer called a rotation speed of 113 knots but the captain did not rotate until 119 knots. The slower than prescribed rotation speed and angle of attack put the aircraft well into the stall regime. Another example of aircraft technologies that were absent on the Fokker 28 4000 are leading edge high lift devices that might have prevented the stall.
The tower cab coordinator on duty at the time of the accident stated that he saw the accident and began rescue efforts. He said that when he tried to use the emergency phone that it was not working properly. He then gave the crash code and location but was unsure if anyone had received it. He then called New York Police department on a separate line. The NTSB investigation revealed that this problem did not significantly hinder rescue efforts but is an area of basic ATC technology that could have been in place to ensure proper emergency response. However the emergency medical response and its coordination significantly did hamper the survivability of this accident. All of the injured persons were not removed until a full one hour and ten minutes after the emergency medical service was notified. A diver was not in the water until 45 minutes after the accident. A contributing factor to this delay was that the Port Authority’s rescue boat could not be launched from its trailer because the construction of a boat lift had not been completed. The rescue attempts had to wait until the U.S. Coast Guard and New York City police boats could join the rescue efforts. Also, the EMS lieutenant reported that no efforts were made to resuscitate victims who appeared to have drowned or lacked vital signs because he felt they had been in the water too long. The NTSB investigation noted that such drowning victims in cold salt water had been successfully resuscitated after periods of up to one hour.
The NTSB found that the probable causes of this accident were the failures of the FAA and the airline industry to provide flight crews with procedures and criteria compatible with departure delays and the decisions of the flight crew to take off without complete assurance that the aircraft’s wings were free of ice contamination. The ice accumulation on the wings caused an aerodynamic stall and loss of aircraft control after lift off. The NTSB also found that a contributing factor to the accident was the inadequate coordination of the flight crew which led to a rotation of the aircraft at slower than appropriate airspeed.
