CASE
12
Aircraft Icing: The Tyranny of Temperature
The aerospace environment is a realm of extremes: low to high pressures, densities, and temperatures. Researchers have had the goal of improving flight efficiency and safety. Aircraft icing has been a problem since the earliest days of flight and, historically, researchers have artfully blended theory, ground-and-flight research, and the use of new tools such as computer simulation and software modeling codes to ensure that travelers fly in aircraft well designed to confront this hazard.
Case-12 Cover Image: Ice formation on aircraft poses a serious flight safety hazard. Here a NASA technician measures ice deposits on a test wing in NASA’s Icing Research Tunnel, Lewis (now Glenn) Research Center, Ohio. NASA.
One February evening in the late 1930s, a young copilot strode across a cold ramp of the Nashville airport under a frigid moonlit sky, climbing into a chilled American Airlines DC-2. The young airman was Ernest Gann, later to gain fame as a popular novelist and aviation commentator, whose best-remembered book, The High and the Mighty, became an iconic aviation film. His captain was Walter Hughen, already recognized by his peers as one of the greats, and the two men worked swiftly to ready the sleek twin-engine transport for flight. Behind them, eight passengers settled in, looked after by a flight attendant. They were bound for New York, along AM-23, an air route running from Nashville to New York City. Preparations complete, they taxied out and took off on what should have been a routine 4-hour flight in favorable weather. Instead, almost from the moment the airliner’s wheels tucked into the plane’s nacelles, the flight began to deteriorate. By the time they reached Knoxville, they were bucking an unanticipated 50-mile-per-hour headwind, the Moon had vanished, and the plane was swathed in cloud, its crew flying by instruments only. And there was something else: ice. The DC-2 was picking up a heavy load of ice from the moisture-laden air, coating its wings and engine cowlings, even its propellers, with a wetly glistening and potentially deadly sheen.[1]
Suddenly there was “an erratic banging upon the fuselage,” as the propellers began flinging ice “chunks the size of baseballs” against the fuselage. In the cockpit, Hughen and Gann desperately fought to keep their airplane in the air. Its leading edge rubber deicing boots, which shattered ice by expanding and contracting, so that the airflow could sweep it away, were throbbing ineffectively: the ice had built up so thick and fast that it shrouded them despite their pulsations. Carburetor inlet icing was building up on each engine, causing it to falter, and only deliberately induced back-firing kept the inlets clear and the engines running. Deicing fluid spread on the propellers and cockpit glass had little effect, as did a hot air hose rigged to blow on the outside of the windshield. Worst of all, the heavy icing increased the DC-2’s weight and drag, slowing it down to near its stall point. At one point, the plane began “a sudden, terrible shudder,” perilously on the verge of a fatal stall, before Hughen slammed the throttles full-forward and pushed the nose down, restoring some margin of flying speed.[2]
After a half hour of desperate flying that “had the smell of eternity” about it, the battered DC-2 and its drained crew entered clear skies. The weather around them was still foreboding, and so, after trying to return to Nashville, finding it was closed, and then flying about for hours searching for an acceptable alternate, they turned for Cincinnati, Hughen and Gann anxiously watching their fuel consumption. Ice—some as thick as 4 inches—still swathed the airplane, so much so that Gann thought, “Where are the engineers again? The wings should somehow be heated.” The rudder was frozen in place, and the elevators and ailerons (controlling pitch and roll) moveable only because of Hughen and Gann’s constant control inputs to ensure they remained free. At dawn they reached Cincinnati, where the plane, burdened by its heavy load of ice, landed heavily. “We hit hard,” Gann recalled, “and stayed earth-bound. There is no life left in our wings for bouncing.” Mechanics took “two hours of hard labor to knock the ice from our wings, engine cowlings, and empennage.” Later that day, Hughen and Gann completed the flight to New York, 5 hours late. In the remarks section of his log, explaining the delayed arrival, Gann simply penned “Ice.”[3]
Gann, ever after, regarded the flight as marking his seasoning as an airman, “forced to look disaster directly in the face and stare it down.”[4]
Many others were less fortunate. In January 1939, Cavalier, an Imperial Airways S.23 flying boat, ditched heavily in the North Atlantic, breaking up and killing 3 of its 13 passengers and crew; survivors spent 10 cold hours in heaving rafts before being rescued. Carburetor icing while flying through snow and hail had suffocated two of its four engines, leaving the flying boat’s remaining two faltering at low power.[5] In October 1941, a Northwest Airlines DC-3 crashed near Moorhead, MN, after the heavy weight of icing prevented its crew from avoiding terrain; this time 14 of 15 on the plane died.[6]
Even when nothing went wrong, flying in ice was unsettling. Trans World Airlines Captain Robert “Bob” Buck, who became aviation’s most experienced, authoritative, and influential airman in bad weather flying, recalled in 2002 that
A typical experience in ice meant sitting in a cold cockpit, windows covered over in a fan-shaped plume from the lower aft corner toward the middle front, frost or snow covering the inside of the windshield frames, pieces as large as eight inches growing forward from the windshield’s edges outside, hunks of ice banging against the fuselage and the airplane shaking as the tail swung left and right, right and left, and the action was transferred to the rudder pedals your feet were on so you felt them saw back and forth beneath you The side winds were frosted, but you could wipe them clear enough for a look out at the engines. The nose cowlings collected ice on their leading edge, and I’ve seen it so bad that the ice built forward until the back of the propeller was shaving it! But still the airplane flew. The indicated airspeed would slow, and you’d push up the throttles for more power to overcome the loss but it didn’t always take, and the airspeed sometimes went down to alarming numbers approaching stall.[7]
Icing, as the late aviation historian William M. Leary aptly noted, has been a “perennial challenge to aviation safety.”[8] It’s a chilling fact that despite a century of flight experience and decades of research on the ground and in the air, today’s aircraft still encounter icing conditions that lead to fatal crashes. It isn’t that there are no preventative measures in place. Weather forecasting, real-time monitoring of conditions via satellite, and ice prediction software are available in any properly equipped cockpit to warn pilots of icing trouble ahead. Depending on the size and type of aircraft, there are several proven anti-icing and de-icing systems that can help prevent ice from building up to unsafe levels. Perhaps most importantly, pilot training includes information on recognizing icing conditions and what to do if an aircraft starts to ice up in flight. Unfortunately the vast majority of icing-related incidents echo a theme in which the pilot made a mistake while flying in known icing conditions. And that shows that in spite of all the research and technology, it’s still up to the pilot to take advantage of the experience base developed by NASA and others over the years.
In the very earliest days of aviation, icing was not an immediate concern. That all changed by the end of the First World War, by which time airplanes were operating at altitudes above 10,000 feet and in a variety of meteorological conditions. Worldwide, the all-weather flying needs of both airlines and military air service, coupled with the introduction of blind-flying instrumentation and radio navigation techniques that enabled flight in obscured weather conditions, stimulated study of icing, which began to take a toll on airmen and aircraft as they increasingly operated in conditions of rain, snow, and freezing clouds and sleet.[9]
The NACA’s interest in icing dated to the early 1920s, when America’s aviation community first looked to the Agency for help. By the early 1930s, both in America and abroad, researchers were examining the process of ice formation on aircraft and means of furnishing some sort of surface coatings that would prevent its adherence, particularly to wings, acquiring data both in actual flight test and by wind tunnel studies. Ice on wings changed their shape, drastically altering their lift-to-drag ratios and the pressure distribution over the wing. An airplane that was perfectly controllable with a clean wing might prove very different indeed with just a simple change to the profile of its airfoil.[10] Various mechanical and chemical solutions were tried. The most popular mechanical approach involved fitting the leading edges of wings, horizontal tails, and, in some cases, vertical fins with pneumatically operated rubber “de-icing” boots that could flex and crack a thin coating of ice. As Gann and Buck noted, they worked at best sporadically. Other approaches involved squirting de-icing fluid over leading edges, particularly over propeller blades, and using hot-air hoses to de-ice cockpit windshields.
Lewis A. “Lew” Rodert—the best known of ice researchers—was a driven and hard-charging NACA engineer who ardently pursued using heat as a means of preventing icing of wings, propellers, carburetors, and windshields.[11] Under Rodert’s direction, researchers extensively instrumented a Lockheed Model 12 light twin-engine transport for icing research and, later, a larger and more capable Curtiss C-46 transport. Rodert and test pilot Larry Clausing, both Minnesotans, moved the NACA’s ice research program from Ames Aeronautical Laboratory (today the NASA Ames Research Center) to a test site outside Minneapolis. There, researchers took advantage of the often-formidable weather conditions to assemble a large database on icing and icing conditions, and on the behavior of various modifications to their test aircraft. These tests complemented more prosaic investigations looking at specific icing problems, particularly that of carburetor icing.[12]
The war’s end brought Rodert a richly deserved Collier Trophy, American aviation’s most prestigious award, for his thermal de-icing research, particularly the development and validation of the concept of air-heated wings.[13] By 1950, a solid database of NACA research existed on icing and its effects upon propeller-driven airplanes.[14] This led many to conclude that the “heroic era” of icing research was in the past, a judgment that would prove to be wrong. In fact, the problems of icing merely changed focus, and NACA engineers quickly assessed icing implications for the civil and military aircraft of the new gas turbine and transonic era.[15] New high-performance interceptor fighters, expected to accelerate quickly and climb to high altitudes, had icing problems of their own, typified by inlet icing that forced performance limitations and required imaginative solutions.[16] When first introduced into service, Bristol’s otherwise-impressive Britannia turboprop long-range transport had persistent problems caused by slush ice forming in the induction system of its Proteus turboprop engines. By the time the NACA evolved into the National Aeronautics and Space Administration in 1958, the fundamental facts concerning the types of ice an aircraft might encounter and the major anti-icing techniques available were well understood and widely in use. In retrospect, as impressive as the NACA’s postwar work in icing was, it is arguable that the most important result of NACA work was the establishment of ice measurement criteria, standards for ice-prevention systems, and probabilistic studies of where icing might be encountered (and how severe it might be) across the United States. NACA Technical Notes 1855 (1949) and 2738 (1952) were the references of record in establishing Federal Aviation Administration (FAA) standards covering aircraft icing certification requirements.[17]
NASA and the Aircraft Icing Gap
At a conference in June 1955, Uwe H. von Glahn, the NASA branch chief in charge of icing research at the then-Lewis Research Center (now Glenn Research Center) in Cleveland boldly told fellow scientific investigators: “Aircraft are now capable of flying in icing clouds without difficulty . . . because research by the NACA and others has provided the engineering basis for ice-protection systems.”[18]
That sentiment, in combination with the growing interest and need to support a race to the Moon, effectively shut down icing research by the NACA, although private industry continued to use Government facilities for their own cold-weather research and certification activities, most notably the historic Icing Research Tunnel (IRT) that still is in use today at the Glenn Research Center (GRC). The Government’s return to icing research began in 1972 at a meeting of the Society of Automotive Engineers in Dallas, during which an aeronautics-related panel was set up to investigate ice accretion prediction methods and define where improvements in related technologies could be made. Six years later the panel concluded that little progress in understanding icing had been accomplished since the NACA days. Yet since the formation of NASA in 1958, 20 years earlier, aircraft technology had fundamentally changed. Commercial aviation was flying larger jet airliners and being asked to develop more fuel-efficient engines, and at the same time the U.S. Army was having icing issues operating helicopters in icy conditions in Europe. The Army’s needs led to a meeting with NASA and the FAA, followed by a July 1978 conference with 113 representatives from industry, the military, the U.S. Government, and several nations. From that conference sparked the impetus for NASA restarting its icing research to “update the applied technology to the current state of the art; develop and validate advanced analysis methods, test facilities, and icing protection concepts; develop improved and larger testing facilities; assist in the difficult process of standardization and regulatory functions; provide a focus to the presently disjointed efforts within U.S. organizations and foreign countries; and assist in disseminating the research results through normal NASA distribution channels and conferences.”[19]
While icing research programs were considered, proposed, planned, and in some cases started, full support from Congress and other stakeholders for the return of a major, sustained icing research effort by NASA did not come until after an Air Florida Boeing 737 took off from National Airport in Washington, DC, in a snowstorm and within seconds crashed on the 14th Street Bridge. The 1982 incident killed 5 people on the bridge, as well as 70 passengers and 4 crewmembers. Only five people survived the crash, which the National Transportation Safety Board blamed on a number of factors, assigning issues related to icing as a major cause of the preventable accident. Those issues included faulting the flight crew for not activating the twin engine’s anti-ice system while the aircraft was on the ground and during takeoff, for taking off with snow and ice still on the airfoil surfaces of the Boeing aircraft, and for the lengthy delay between the final time the aircraft was de-iced on the tarmac and the time it took the crew to be in position to receive takeoff clearance from the control tower and get airborne. While all this was happening the aircraft was exposed to constant precipitation that at various times could be described as rain or sleet or snow.[20]
The immediate aftermath of the accident—including the dramatic rescue of the five survivors who had to be fished out of the Potomac River—was all played out on live television, freezing the issue of aircraft icing into the national consciousness. Proponents of NASA renewing its icing research efforts suddenly had shocking and vivid proof that additional research for safety purposes was necessary in order to deal with icing issues in the future. Approval for a badly needed major renovation of the IRT at GRC was quickly given, and a new, modern era of NASA aircraft icing investigations began.[21]
Baby, It’s Cold Out There
Not surprisingly, ice buildup on aircraft is bad. If it happens on the ground, then pilots and passengers alike must wait for the ice to be removed, often with hazardous chemicals and usually resulting in flight delays that can trigger a chain reaction of schedule problems across the Nation’s air system. If an aircraft accumulates ice in the air, depending on the severity of the situation, the results could range from mild annoyance that a de-icing switch has to be thrown to complete aerodynamic failure of the wing, accompanied by total loss of control, a spiraling dive from high altitude, a premature termination of the flight and all lives on board, followed by the reward of becoming the lead item on the evening news.
Icing is a problem for flying aircraft not so much because of the added weight, but because of the way even a tiny amount of ice can begin to disrupt the smooth airflow over the wings, wreaking havoc with the wing’s ability to generate lift and increasing the amount of drag, which slows the aircraft and pitches the nose down. This prompts the pilot to pull the nose up to compensate for the lost lift, which allows even more ice to build up on the lower surface of the wing. And the vicious circle continues, potentially leading to disaster. Complicating the matter is that even with options for clearing the wing of ice—discussed shortly—ice buildup can remain and/or continue on other aircraft surfaces such as antennas, windshields, wing struts, fixed landing gear, and other protrusions, all of which can still account for a 50-percent increase in drag even if the wing is clean.[22]
From the earliest experience with icing during the 1920s and on through the present day, researchers have observed and understood there to be three primary categories of aircraft ice: clear, rime, and mixed. Each one forms for slightly different reasons and exhibits certain properties that influence the effectiveness of available de-icing measures.[23]
Clear ice is usually associated with freezing rain or a special category of rain that falls through a region of the atmosphere where temperatures are far below the normal freezing point of water, yet the drops remain in a liquid state. These are called super-cooled drops.
A graphic depicting clear ice buildup on an airfoil.
Such drops are very unstable and need very little encouragement to freeze. When they strike a cold airframe they begin to freeze, but it is not an instant process. The raindrop freezes as it spreads out and continues to make contact with an aircraft surface whose skin temperature is at or below 32 degrees Fahrenheit (0 degrees Celsius). The slower the drop freezes, the more time it will have to spread out evenly and create a sheet of solid, clear ice that has very little air enclosed within. This flow-back phenomenon is greatest at temperatures right at freezing. Because of its smooth surface, clear ice can quickly disrupt the wing’s ability to generate lift by ruining the wing’s aerodynamic shape. This type of ice is quite solid in the sense that if any of it does happen to loosen or break off, it tends to come off in large pieces that have the ability to strike another part of the aircraft and damage it.[24]
A graphic depicting rime ice buildup on an airfoil.
Rime ice proves size makes a difference. In this case the super-cooled liquid water drops are smaller than the type that produces clear ice. When these tiny drops of water strike a cold aircraft surface, most of the liquid drops instantly freeze and any water remaining is not enough to create a sheet of ice. Instead, the result is a brittle ice that looks milky white, is opaque, has a rough surface due to its makeup of ice crystals and trapped air, and doesn’t accumulate as quickly as clear ice. It does not weigh as much, either, and tends to stick to the leading edge of the wing and the cowl of the engine intakes on a jet, making rime ice just as harmful to the airflow and aerodynamics of the aircraft.[25]
Naturally, when an aircraft encounters water droplets of various sizes, a combination of both clear and rime ice can form, creating the third category of icing called mixed ice. The majority of ice encountered in aviation is of this mixed type.[26]
A graphic depicting mixed ice buildup on an airfoil.
Aircraft must also contend with snow, avoiding the wet, sticky stuff that makes great snowballs on the ground but in the air can quickly accumulate not only on the wings—like ice, a hazard in terms of aerodynamics and weight—but also on the windshield, obscuring the pilot’s view despite the best efforts of the windshield wipers, which can be rendered useless in this type of snow. And on the ground, frost can completely cover an aircraft that sits out overnight when there is a combination of humid air and subfreezing temperatures. Frost can also form in certain flying conditions, although it is not as hazardous as any of the ices.[27]
Melting Your Troubles Away
As quickly as the hazards of aircraft icing became known in the early days of aviation, inventive spirits applied themselves to coming up with ways to remove the hazard and allow the airplane to keep flying. These ideas at first took the form of understanding where and when icing occurs and then simply not flying through such conditions, then ways to prevent ice from forming in the first place—proactive anti-icing—were considered, and at the same time options for removing ice once it had formed—reactive de-icing—were suggested and tested in the field, in the air, and in the wind tunnel. Of all the options available, the three major ones are the pneumatic boot, spraying chemicals onto the aircraft, and channeling hot bleed air.[28]
A King Air equipped with a de-icing boot on its wing leading edge shows how the boot removes some ice, but not on areas behind the boot.
The oldest of the de-icing methods in use is the pneumatic boot system, invented in 1923 by the B.F. Goodrich Corporation in Akron, OH. The general idea behind the boot has not changed nearly a century later: a thick rubber membrane is attached to the leading edge of a wing airfoil. Small holes in the wing behind the boot allow compressed air to blow through, ever so slightly expanding the boot’s volume like a balloon. Any time that ice builds up on the wing, the system is activated, and when the boot expands, it essentially breaks the ice into pieces, which are quickly blown away by the relative wind of the moving aircraft. Again, although the general design of the boot system has not changed, there have been improvements in materials science and sensor technology, as well as changes in the shape of wings used in various sizes and types of aircraft. In this manner, NASA researchers have been very active in coming up with new and inventive ways to enhance the original boot concept and operation.[29]
One way to ensure there is no ice on an aircraft is to remove it before the flight gets off the ground. The most common method for doing this is to spray some type of de-icing fluid onto the aircraft surface as close to takeoff as possible. The idea was first proposed by Joseph Halbert and used by the United Kingdom Royal Air Force in 1937 on the large flying boats then operated by Imperial Airways.[30] Today, the chemicals used in these fluids usually use a propylene glycol or ethylene glycol base and may include other ingredients that might thicken the fluid, help inhibit corrosion on the aircraft, or add a color to the mixture for easier identification. Often water is added to the mixture, which although counterintuitive makes the liquid more effective. Of the two glycols, propylene is more environmentally friendly.[31]
The industry standard for this fluid is set by the aeronautics division of the Society of Automotive Engineers, which has published standards for four types of de-icing fluids, each with slightly different properties and intentions for use. Type I has a low viscosity and is usually heated and sprayed on aircraft at high pressure to remove any snow, ice, or frost. Due to its viscosity, it runs off the aircraft very quickly and provides little to no protection as an anti-icing agent as the aircraft is exposed to snowy or icy conditions before takeoff. Its color is usually orange.[32] Type II fluid has a thickening agent to prevent it from running very quickly off the aircraft, leaving a film behind that acts as an anti-icing agent until the aircraft reaches a speed of 100 knots, when the fluid breaks down from aerodynamic stress. The fluid is usually light yellow. Type III fluid’s properties fall in between Type I and II, and it is intended for smaller, slower aircraft. It is popular in the regional and business aviation markets and is usually dyed light yellow. Type IV fluids are only applied after a Type I fluid is sprayed on to remove all snow, ice, and frost. The Type IV fluid is designed to leave a film on the aircraft that will remain for 30 to 80 minutes, serving as a strong anti-icing agent. It is usually green.[33]
