Safety areas for airplanes that overshoot or undershoot runways at National Airport are too short to meet current federal standards, but the airport's crash and rescue preparations go beyond what is required, a federal inquiry into the crash of Air Florida Flight 90 was told yesterday.
Hearings into the Jan. 13 crash yesterday turned to layout and safety conditions at National Airport, where the Boeing 737 began its 30-second flight that ended when it struck the 14th Street bridge and plunged into the Potomac River, killing 78 people.
National's operator, the Federal Aviation Administration, says that the airport is fully safe. But over the years, pilots and civic groups have argued the facility's short runways and overrun areas and its curved approach route make it significantly less safe than newer, larger airports like Dulles International.
Yesterday, FAA certification official Carl Steinhauer testified that to receive approval, any airport built after 1970 generally must have 1,000 feet of level ground beyond its runways to accommodate planes that undershoot or run off the end.
When National opened in 1941, before the era of high-speed jet aircraft, the rules called only for 200-foot overruns. Due to grandfather clauses, those standards continue to apply to the airport.
Four of National's six runway terminations fail to meet the 1941 standard but have been exempted. In 1972, the FAA granted an exemption request from the airport manager, who argued that extending runways into the Potomac was not feasible.
Flight 90 took off from Runway 18/36, which is 6,870 feet long and has a 335-foot overrun area at its northern end. Newer airports, such as Dulles International, have runways extending 11,000 feet with 1,000-foot overruns.
Some investigators have speculated that Flight 90's pilots wanted to abort their take-off but, fearing the plane would skid off the snowy runway's end and slide into the Potomac, elected to try to get the plane off the ground instead.
The National Transportation Safety Board, which is directing the crash inquiry, has argued for years that more airports should be required to build 1,000-foot overruns and that certification standards should be tightened. The FAA is studying these issues but has contended that current standards provide for sufficient safety.
Steinhauer also said that National Airport's preparations for a crash or fire exceed those required by federal regulations. The airport is required to have three firefighting vehicles, investigation documents show, but has five on call and one in reserve.
Federal rules do not require that the airport's firefighters respond to crashes off its property or in the Potomac, he said. On the day of the accident, National had a fan boat and a pontoon boat on call. Federal rules do not require "mutal aid" agreements with rescue officials in surrounding jurisdictions, but National has them, he said.
FAA inspectors found no deficiencies at National that rated as "serious" in 1981, Steinhauer said. In 1980, documents show, they found one such deficiency, the alteration of a private company's fuel trucks in a manner the FAA considered to be dangerous.
In another development, safety board member Francis McAdams announced that the FBI would be asked to conduct electronic voice analysis on a part of Flight 90's cockpit voice recorder tape. The purpose is to clear up disagreement as to whether one of the pilots responded during the pre-flight checklist that engine anti-icing devices were "on" or "off."
McAdams said the consensus of those who have listened to the tape, which has not been released to the public, was that the pilot said "off." But he agreed to further analysis in response to protests from Air Florida, the Air Florida Pilots Association and the Air Line Pilots Association that the pilot might have said "on."
Engine anti-icing is an important issue in the crash investigation, due to theories that the pilots did not turn the devices on, leading thrust sensors in the engines to freeze. This created erroneous thrust readings in the cockpit, the theory goes, which led the pilots to set their throttles too low.
Yesterday, investigators also questioned a safety board technician who analyzed engine whine on the plane's cockpit voice recorder tape. His analysis showed that as Flight 90 began its take-off roll, engine speed rose, then decreased slightly. That bolstered speculation that the pilots initially set the throttles correctly, then trimmed them back to false thrust readings. About 10 seconds before impact, analysis of the tape shows, full power was applied.
Yesterday, the board released new information on another suspected factor in the crash, the known tendency of Boeing 737s to pitch up at the nose or roll unexpectedly in icy conditions, creating serious control problems. Previously, investigators knew of some half dozen incidents but a list released yesterday showed 13 incidents reported to Boeing between 1970 and last winter and another 10 reported to Boeing or in the media since the Air Florida crash.
Boeing has issued two bulletins to airlines warning of the control problem, published articles about it in its aviation magazine and conducted studies into its cause. But the circumstances of the Air Florida accident--take-off in icy conditions and use of thrust reversers on the ground, which can throw ice onto a plane's wings--closely match those in many of the reported incidents.