Comments submitted to the docket in response to the FAA’s
proposal for new icing certification regulations range from self-serving promotion of new inventions to carefully considered suggestions about how to improve icing safety after decades of study and hundreds of fatalities.

The comments submitted by former FAA engineer and icing expert John Dow are clearly focused on the safety issues raised by the FAA’s notice of proposed rulemaking (NPRM). Dow points out in his submission that even if the new rules make it through the agency’s cumbersome rulemaking process on schedule, the effects will not be realized until 25 years since the American Eagle Flight 4184 accident that precipitated the FAA’s action. The accident occurred near Roselawn, Ind., on Oct. 31, 1994, killing 68 passengers and crew in an ATR 72. “The time period for the aviation industry to propose this amendment that they must comply with and for the FAA to bring it to the NPRM stage has been 15 years, seven months, 28 days from the date of the accident,” Dow noted. “If [it is] codified and effective at the beginning of say, 2011, transport airplanes will have five years for a new design to be certified to this rule or 2016 before an airplane would enter service and several more years for a fleet to accrue significant exposure.”

While the NPRM explores expansion of the icing conditions envelope that manufacturers would have to address during certification for Part 25 transport-category aircraft, Dow is concerned that the FAA is neglecting operational factors that could have a much greater effect on safety. In any case, the new rules wouldn’t take effect for a few more years and would apply only to new Part 25 airplanes and not Part 23 airplanes. Oddly, as Dow explained, the NPRM even mentions that Part 23 airplanes are “more critical with respect to icing and are addressed in the cost/benefit analysis; however, no further mention is made [in the NPRM] with reference to small and commuter-category airplanes regulated under...Part 23.”

The key information for pilots in this potentially confusing process is, fortunately, clearly laid out in Dow’s comments. This, more than any rulemaking that parses tiny diameters of supercooled liquid water droplets, could have a much more far-reaching and beneficial effect on icing safety. As Dow says in his comments, “Even though [post-stall recovery practical test standards] were not addressed in the NPRM, they are relevant and should not be procedurally dismissed.”  

According to Dow, “A common pilot input characteristic in several fatal icing-related accidents has been observed on the flight data recorder time histories. The common pilot input to the controls has been to add power and maintain the pitch angle of the airplane and in some cases apply nose-up pitch input even against the nose-down force being applied by the airplane’s ‘stick pusher’ that is designed to rapidly reduce the angle of attack! Especially in icing, but not exclusively, this practice has delayed or made recovery impossible before ground contact. These habit patterns are developed and reinforced as the required response in simulator training in accordance with FAA practical test standards for stall identification and recovery for minimum altitude loss. ‘Minimum altitude loss’ is trained as ‘zero altitude loss.’”

In aircraft with stick shakers or pushers and those without, the only way to recover from an icing-induced stall is to reduce the angle of attack, not by adding power and trying to maintain attitude, according to Dow. “It is the only way to prevent loss of control and a spin in an iced airplane.” This technique, he added, could easily have prevented the Roselawn accident, the Monroe, Mich., accident (Embraer EMB-120RT, Jan. 9, 1997, 29 dead), the Buffalo accident (Bombardier Q400, Feb. 12, 2009, 50 dead) and an accident in Crezzo, Italy (Oct. 15, 1987, 37 dead).
Dow has pitched the idea of changing the FAA’s focus on “hold pitch and add power” to reducing “angle of attack by nose-down control input,” but he wrote that the FAA dismissed this suggestion. He reiterated the concept in an April 2005 article in the Flight Safety Digest, “Understanding the Stall Recovery Procedure for Turboprop Airplanes in Icing Conditions.” (The article is available online; type the title into a search engine.) Pilots need to be taught, especially in simulator training, to recognize the onset of aerodynamic stalls, not by artificial stall warning cues but by “g” break, buffet or roll oscillations, he explained. “Current FAA efforts to change the [Practical Test Standards] must be expanded in a more expeditious manner to allow stalls to be injected into the simulator training at any point in the flight in lieu of a highly scripted, predictable and unrealistic procedure.”

This knowledge, sadly, is not new, and Dow notes that lessons learned trace as far back as the March 31, 1931 crash of a Fokker F10, the accident in which Notre Dame football coach Knute Rockne died. “This was the perhaps the first major accident of national interest involving freezing drizzle/freezing rain,” Dow wrote. Unfortunately the investigation focused on the failure of the wing structure following the pilot’s attempt to recover from an upset instead of the cause of the upset. “Hopefully, after almost 80 years since the first well publicized freezing drizzle/freezing rain accident in the U.S., 75 years since a procedure to identify susceptibility to ridge ice and ice protection system limits and at least 25 years after Roselawn, a truly effective regulation will be codified. It might even occur before mankind sets foot on Mars,” he wrote.

In its comments about the NPRM, the NTSB noted that its recommendation to the FAA that icing certification regulations be revised “to ensure that airplanes are properly tested for all conditions in which they are authorized to operate, or are otherwise shown to be capable of safe flight into such conditions” remain open with an unacceptable response. The Safety Board added that it “is pleased that the NPRM proposes to add Section 25.1420, which would require evaluating the operation of airplanes in the SLD environment, and Appendix O, Part I, to Part 25, which expands the certification icing environment to include freezing rain and freezing drizzle by using four separate droplet size distributions and includes droplet sizes greater than 1,000 microns in one of the distributions.”

The Board agrees with Dow that the NPRM neglects Part 23 airplanes, which are far more susceptible to the dangers of icing conditions. “The NTSB is disappointed with the proposed rule and believes that it should be expanded to include all aircraft regardless of maximum takeoff weight or flight control design and that the requirements should apply to both newly manufactured and currently certificated aircraft under Part 25 and Part 23.”

The European Aviation Safety Agency submitted comments and highlighted one area that the NPRM doesn’t address, airspeed indicating system malfunctions. While the NPRM does mention mixed phase icing and ice crystals affecting pitot tubes and angle of attack sensors, the EASA recommends “having a specific requirement for flight instrument external probes, including but not necessarily limited to pitot, pitot-static and static probes, alpha vanes, side-slip vanes and temperature probes.” Additionally, the EASA wants Part 25.1326 (pitot heat indication systems) modified “to extend the scope of the requirement to all flight instrument external probes, including but not necessarily limited to pitot, pitot-static and static probes, alpha vanes, side-slip vanes and temperature probes.”

NBAA Offers Icing Webinar
To help pilots prepare for the winter flying season in the northern hemisphere, NBAA held a webinar titled “Icing and Winter Operations: Review the Essentials” on September 9. The webinar is available on the NBAA Web site for $99 (members) and $149 (nonmembers), in the professional development section. The webinar is presented by Judith Van Zante, a Ph.D in aerospace engineering from the University of Southern California and contract employee at NASA’s Glenn Research Center in Cleveland. Van Zante has flown as a flight-test engineer on the NASA/FAA tailplane icing research program and also led the team that produced NASA’s online icing training aids for pilots.

The webinar covers key areas (also available in the NASA online training courses), including risks of contamination, preflight planning, ground de/anti-icing operations, in-flight icing and the important new subject of turbine engine and ice crystal icing.