Aviation Week Takes A NASA Core Icing Research Flight | Technology content from Aviation Week
"Aviation Week Takes A NASA Core Icing Research Flight
Researchers fly into tropical storms to evaluate possible warning methods for core icing events
The warm subtropical waters of the Gulf of Mexico in late summer may seem an unusual place to hunt for a rare form of atmospheric icing, but this is prime research territory for NASA and the agency’s highly instrumented DC-8 aircraft.
Searching for super-cold conditions from the sweltering heat of a Florida airfield is as counterintuitive as the ice crystal icing (ICI) phenomenon NASA is trying to find, a condition in which ice particles accumulate inside the hot core of a jet engine. Also known as high ice water content (HIWC), the state occurs without warning, generally at high altitudes above normal icing levels, and can result in temporary power loss, surges, blade damage and, in some severe cases, engine shutdown.
Researchers believe that ice crystals start to melt and evaporate as they meet warm parts inside the engine, cooling core surfaces to temperatures below freezing. The cooling engine causes the melted ice crystal water to refreeze, and ice accumulates inside the engine core. At some point, slabs of ice come loose and are ingested, causing power loss or blade damage.
While the dangers of traditional icing at medium altitudes are well understood and easily countered, the high-altitude HIWC scenario continues to puzzle researchers. The number of core icing events appears to have mushroomed from virtually nothing to more than 150 known incidents over the past two decades, and the rate is increasing. The growing incidence of core icing has forced changes in aircraft operating procedures and prompted the creation of a new set of certification standards for engines and avionics.
Solving the HIWC mystery is important. Researchers theorize that ice core incidents are on the rise partly because more airliners are flying with greater frequency through mid-latitude and subtropical regions prone to intense convection. In addition ice crystals can affect aircraft data systems leading to errors in readings of temperatures, air speed and angle of attack. Icing is thought to have contributed to the loss of Air France 447, which crashed into the Atlantic after flying through storms in 2009.
Two international HIWC research groups are focused on tackling the icing problem. The North American HIWC study group involves NASA, the FAA, Transport Canada and Environment Canada, Airbus, Boeing and the Australia Bureau of Meteorology. The European high-altitude ice crystals (HAIC) consortium, which is coordinated by Airbus, brings together 34 industrial and research partners from 11 European countries and five from Australia, Canada and U.S. Both groups are coordinating research on three main objectives; understanding the physics of the HIWC process, developing new regulatory guidelines and developing HIWC detection methods.
Aviation Week was invited to join a NASA HIWC research flight from Fort Lauderdale, Florida. The test campaign forms part of what is “really a three-pronged approach to cracking this nut,” says Ron Colantonio, project manager for engine icing research at NASA Glenn Research Center, Ohio. “First we have to characterize the weather that’s causing the problem. Are the ice crystals small or big? How much is the water concentration? We don’t have a handle on things like that,” he says.
“Second, we want to know what’s happening in the engine. What are the physics? It’s not intuitive you can have icing inside an engine.” To reach these answers, and to develop a simulation tool for engine makers to use for testing designs against core icing, Glenn has modified its Propulsion Systems Laboratory (PSL) to replicate HIWC conditions. The facility has been tested with a fully instrumented Honeywell LF502, which was one of the earliest high-bypass turbofans to exhibit vulnerability to “roll back” caused by icing.
“We can ‘fly’ an engine from the ground up to 40,000 ft. The ice crystals are generated in front by spraying liquid water into a cold airstream. By the time it gets to the engine it is frozen,” says Colantonio. “It is the only capability in the world that can do this, and our hope is with this flight campaign we can see we are indeed replicating the right conditions in this facility. We think this is the new method of ground-based engine ice testing, and we are trying to develop new test methodologies and new ways to calibrate the facility. We need new diagnostics for ice crystals and water content. We are on the right path and we should have a solid capability in the next few years,” he adds.
With updated engine certification regulations on the way to cover core icing requirements, NASA is working with the FAA and other agencies to use the PSL as a means of compliance. “We have an icing research tunnel that manufacturers could come to from around the world,” says Colantonio.
The third target is to develop detection and warning methods, preferably by adapting existing onboard equipment. “The focus of these flights is to see if we can use remote-sensing capabilities, like the weather radar [to detect and avoid HIWC]. It is like the low-hanging fruit,” he adds. For the HIWC program the DC-8 was fitted with a Honeywell RDR-4000 X-band weather radar, which is designed to measure liquid water content and turbulence. “A lot of the signal is filtered and we are going into engine icing conditions and saving the raw radar data; we hope to see that we can detect ice crystals with the radar or at least infer if there are ice crystals there,” says Colantonio.
One puzzling question researchers hope to answer is why existing radars routinely fail to pick up ICI-like conditions at all, despite the very high water content. Air crews frequently report that weather radar at the time of engine thrust loss shows relatively benign green (low reflectivity) or even black, indicating no discernible threat. NASA Langley Research Center weather radar principal investigator Steven Harrah says: “The amount of moisture that’s deemed to be in these HIWC clouds is 3 grams/cu. meter or more, and if you converted that into precipitated rain, it would be 2 in. per hour, which is heavy rain. On a weather radar, that’s deep into the red, so we are scratching our heads and saying, ‘why aren’t we seeing this already?’ The systems work perfectly, so where’s the missing part?” The answer likely boils down to droplet size, says Harrah, but even the variations found within a typical convective cell or between different regions of the world cannot account for the discrepancy.
A potential answer is that the drop size distribution is skewed, and that although there is a lot of water held within the cloud, it is made up predominantly of particles too small to be picked up the radar. Weather radars operate on wavelengths optimized for around 1 mm, “which works very well, but for HIWC conditions that are maybe dominated by particles of 100 microns or smaller, that’s not so good,” he adds.
While dual-band radar could provide one answer, the preferred solution likely will be the less expensive route of adapting current radars with new signal-processing algorithms that will be able to “infer” the presence of HIWC. “We can measure convection and reflectivity in the atmosphere, where there is lot of moisture, so we can measure a high probability of HIWC at a certain point in space. That’s just a software change in the existing box,” says Harrah.
Specifically, researchers believe the location of the HIWC event can be inferred by bringing together evidence from the existing data or which can be gleaned from new modes. Convection can be mapped using a NASA-developed mode that measures vertical winds. Higher reflectivity at lower altitudes—indicating higher visible water amounts around the convective area—is another clue, while radial winds that advect moisture away from the convection “chimney” will also help locate the danger area. “Looking at those winds, we can infer where it should be and using outside air temperature and other parameters we can fine-tune it,” says Harrah. Preventing nuisance alerts will be key. “If pilots turn it off, that’s the worst possible result,” he adds.
For the test campaign, which included 10 flights over 20 days, the goal was to record both instrumented weather and standard radar data as the DC-8 flew in known HIWC conditions, and then see if by comparing the data a potential HIWC radar signature could be identified. The instrument suite included an isokinetic evaporator probe (IKP), a pitot-style forward-scattering device that measures ice water content, and three devices to gather data on particle shape and size. The trio was made up of a cloud droplet probe for particles 2-50 microns, a particle-imaging probe (PIP) for 100 microns to around 6.2 mm, and a 2DS stereoscopic probe for 10 microns to 1.2 mm.
“We’re trying to relate these ice water content-level measurements to what we what radar signatures are being recorded,” says Tom Ratvasky, in-situ probes co-principal investigator for NASA Glenn. “The data we are collecting and the technology we are working on for a more sensitive radar could be offered to assist the current fleet to avoid hazards. It will also help development of real-time nowcasting tools for detecting HIWC conditions, which the FAA is very interested in. The flights will also help with pilot education,” he notes.
For the HIWC campaign, the normally NASA Armstrong Flight Research Center-based DC-8 was temporarily housed on Florida’s Atlantic Coast. From here, researchers had the option of targeting convective weather over the Caribbean and western Atlantic or nearby Gulf of Mexico. On the day of Aviation Week’s 7-hr. flight, a line of thunderstorms was marching across the gulf and intensifying over Louisiana, providing promising conditions for engine icing.
Closing on the line of storms from the east, the aircraft was flown initially on the west side of the system’s stratiform clouds at 37,000 ft. Turning north, the DC-8 was flown through the edges of the storm, avoiding the areas of maximum turbulence red-painted on the radar just as any commercial airliner crew would do. Careful measurements of water content, droplet size, radar data, air temperature and other parameters continued throughout the flight, as the aircraft made several passes around a north-south oriented track through the storm. The aircraft then descended, and the pattern of observation was repeated at 34,000 ft. and 29,000 ft.
Good HIWC data was successfully gathered, and ice buildup on the air data and total air temperature probes around the cockpit was noted by the crew, even though the radar was indicating green or black. Several crews experiencing ice core events have also reported temperature anomalies, and for the first time on any research flight, the NASA campaign recorded total air temperature probe indication changes from well below freezing to freezing, and the temporary failure of the aircraft’s pitot tubes. Later in the campaign, the DC-8 also flew into Tropical Storms Danny and Erike, the first time that an aircraft equipped with both an ice water measurement suite and pilot weather radar was able to record conditions associated with such events."