The FAA has been giving a great deal of thought to procedures for coping with icing encountered while an airplane sits on the ground, and before the winter flying season begins, we can expect to see new rules governing how air carriers address this important safety issue. The research behind these rules could pay long-term dividends to general aviation as well.
The NTSB counts 15 air carrier accidents over the past 23 years relating to failure to adequately deice an airplane before takeoff. Seven of these involved major carriers engaged in FAR Part 121 operations. Five involved passenger-carrying flights that resulted in 135 fatalities and 66 serious injuries. (Five also involved early model Douglas DC-9s, which should tell someone something.)
Several of these accidents are notorious both inside and outside the aviation industry, including the November 15, 1987, crash of a Continental DC-9 in Denver that killed 25 passengers, the pilots, and a flight attendant, and the January 13, 1982, crash of an Air Florida Boeing 737 attempting a takeoff from Washington National that resulted in the death of 74 of the 79 people aboard and four people on the ground.
On July 23, the FAA released a notice of proposed rulemaking (NPRM 92-9) that would require Part 121 certificate holders "to develop an FAA-approved ground deicing/anti-icing program and to comply with that program any time conditions are such that frost, ice, or snow could adhere to the aircraft's wings, control surfaces, propellers, engine inlets, and other critical surfaces."
The accident that apparently motivated the NPRM is not counted in the numbers cited above: It was the crash at New York's La Guardia Airport of a USAir Fokker F-28 on March 22 of this year. While the NTSB had not, at the time of this writing, issued a probable cause finding for this accident, "the FAA is proceeding on the assumption that the accident was caused, at least in part, by icing," according to the NPRM. The grounds for this conclusion — and for the flurry of FAA activity — could rest, at least in part, on the widely reprinted comments of some passengers on that flight regarding wing contamination they observed before the attempted takeoff. The traveling public, it appears, has become sensitized to the issue.
Also in response to the USAir accident, the FAA convened an International Conference on Airplane Ground Deicing on May 28 and 29 in Reston, Virginia, to share information on ground deicing/anti-icing and to recommend actions for preventing icing accidents.
According to the NPRM, "Two major recommendations made by the working groups that support this rulemaking are: (1) Critical aircraft surfaces must be kept free of frost, ice, and snow; and (2) Each air carrier should have an approved aircraft deicing program that will assure full compliance with the clean aircraft concept. The program should include ground deicing, a comprehensive training program for flight crewmembers, holdover timetables to be used as guidelines [ holdover time is the estimated time deicing or anti-icing will prevent the formation of frost or ice and the accumulation of snow or slush on the treated surfaces of an aircraft — Ed.], and criteria for determining if a pretakeoff inspection after deicing is needed. (There was no consensus on when a pretakeoff inspection must be conducted.)"
The upshot of all this for general aviation is that improved methods, materials, procedures, educational aids, and flight- and ground-personnel training could greatly enhance the safety of GA operations at a time when more pilots are becoming instrument qualified and more flights are being made in inclement weather.
Safety pundits have remarked for years that, when ice is encountered, the person at the controls of the aircraft becomes a test pilot. Were there any doubt this is true, reflect on comments made by the deputy chief design engineer of the DC-9 program in 1988:
"For an airplane trimmed for takeoff, the stabilizer is set to balance the moments due to both aerodynamic forces and center of gravity location so that the stick force at climb-out speed ranges from none to a slight pull. This balance is upset by wing ice contamination, particularly on contemporary aircraft with tapered, swept wings. With contamination on the wings, the aircraft will increasingly behave as if it was mistrimmed in the airplane nose-up direction as the angle of attack is increased. This will result in the airplane's pitching up more rapidly than normal during the takeoff rotation, and will require an abnormal push force to maintain the desired airspeed during climb."
Following the February 17, 1991, crash of a Ryan International Airlines DC-9 freighter attempting to take off from Cleveland-Hopkins (Ohio) International, which killed both crewmembers, McDonnell Douglas's vice president for flight operations/labs/safety and training, T. M. Ryan, Jr., sent a letter to DC-9 operators that reiterated some of the earlier report's major points and offered some observations valuable to pilots of all types of airplanes:
"Ice contamination adversely affects (1) straight-wing aircraft such as the Nord 262 and numerous general aviation aircraft, (2) small turbojet aircraft with conventional airfoils (i.e., no leading-edge high-lift devices) such as the Learjet, (3) larger aircraft with conventional airfoils such as the F-28, DC-9-10, and DC-8, and (4) aircraft with leading-edge high-lift devices such as the 737. In most takeoff accidents, the ice contamination has not been large ice accretions on the leading edges, or thick layers of adhering snow on top of the wings. Rather, dangerous reductions in handling qualities and stall margins can occur because of icing roughness equivalent to that of medium-grit sandpaper. This seemingly modest amount of contamination can result in pitching moment changes during takeoff rotation that cause the airplane to increasingly behave as if it were mistrimmed in the nose-up direction. Following liftoff, degraded lateral stability requires larger and larger control wheel inputs to keep the airplane from abruptly rolling off, possibly followed by premature stall at lower than normal angles of attack. Additionally, the airflow into the engines may become disturbed, causing compressor surges and momentary losses of power.
"As might be expected, the leading edge portion of the wing and the wing upper surfaces are the most sensitive to surface roughness, such as that caused by ice contamination. Ice accumulation on the wing surface is very difficult to detect. It cannot be seen from ahead of the wing during walkaround, is very difficult to see from behind the wing, and may not be detectable from the cabin because it is often clear and wing surface details may show through.
"These contaminants produce three major aerodynamic effects:
"1. When operating in the low-speed regime common to takeoff and final approach, the stall margins at the target airspeeds are substantially reduced.
"2. For a given angle of attack, the wing produces less lift and therefore requires higher pitch attitudes (and/or higher speeds) to achieve liftoff.
"3. The angle of attack at the point of stall is reduced to below that of an uncontaminated wing, and may cause the stall to occur before stall warning devices activate.
"...Scrupulously careful ice inspections shortly before takeoff are a must whenever atmospheric conditions make it prudent to do so. Even suspicious conditions justify inspection or precautionary deicing. Crews should be encouraged to taxi back for a second deicing if a delayed takeoff in freezing precipitation raises any question of wing condition...."
While light general aviation airplanes seldom have tapered, swept wings, their pilots face the same dilemma of decreased stall angle of attack, decreased lift (in response to which the temptation is to increase the angle of attack), and the potential of decreased thrust due to icing of engine air inlets and propellers. An accident is all but inevitable when the angle of attack required to keep the airplane aloft is greater than the stall angle of attack, particularly if lateral stability deteriorates.
"Until recently," NPRM 92-9 states, "the FAA and the aviation community in general had placed priority on emphasizing the need during icing conditions for the pilot in command to ensure 'clean wings' before takeoff. The FAA believed that pilot education appeared key to combating the threat of wing icing. Although the FAA still believes the pilot in command must ultimately make the decision on whether to take off, and that the decision must be based on a thorough understanding of factors involved in icing, the FAA has determined that the certificate holder must provide the pilot in command with criteria on which to make a proper decision. This proposed rule would require that the pilot in command be provided with information to assist the pilot in determining if the aircraft is free of contamination before takeoff."
The final rule arising from this NPRM, expected to become effective on November 1, will be costly for Part 121 carriers and have unforeseeable effects on flight delays during periods of bad weather. The standards it embraces, however, could significantly advance the level of knowledge and the state of the art of deicing/anti-icing technology available to the general aviation pilot.
We'll take a closer look at deicing/anti-icing procedures as applied to GA in an upcoming "By the Book."
Related Articles
