Document
-9, a3- q I
t::- 63 i-S
NASA Technical Memorandum 104518
NASA's Aircraft Icing
Technology Program
John J. Reinmann
Lewis Research Center
Cleveland, Ohio
Prepared for the
1991 Winter Annual Meeting of the American
Society of Mechanical Engineers
Atlanta, Georgia, December 1-6, 1991
NASA
NASA's AIRCRAFT ICING TECHNOLOGY PROGRAM John J. Reinmann National Aeronautics and Space Administration Lewis Research Center Cleveland, Ohio 44135 SUMMARY References 1 to 3 review major elements of NASA's new icing technology program, up to about mid-1989.
NASA's Aircraft Icing Technology program is aimed Reference 4 provides a selected bibliography of the at developing innovative technologies for safe and aircraft icing work done between 1939 and 1955 by NACA efficient flight into forecasted icing. The program (predecessor to NASA). Reference 5 surveys the state- addresses the needs of all aircraft classes and sup- of-the-art, internationally, in computer icing simula- ports both commercial and military applications. The tion and experimental icing simulation.
program is guided by three key strategic objectives: The purposes of this paper are to review the stra- (1) numerically simulate an aircraft's response to an tegic objectives for NASA's icing program, to present in-flight icing encounter, (2) provide improved experi- selected results obtained since 1989 that illustrate mental icing simulation facilities and testing tech- progress toward these objectives, and to provide an niques, and (3) offer innovative approaches to ice updated bibliography of work supported by NASA and its protection. Our research focuses on topics that collaborators.
directly support stated industry needs, and we work The charter for NASA's aircraft icing technology closely with industry to assure a rapid and smooth program is to develop innovative technologies for safe transfer of technology. This paper presents selected and efficient flight into forecasted icing. The pro- results that illustrate progress toward the three stra- gram addresses the needs of all aircraft classes and tegic objectives, and it provides a comprehensive list supports both commercial and military applications.
of references on the NASA icing program. Our research focuses on topics that directly support stated industry needs, and we work closely with INTRODUCTION industry to assure a rapid and smooth transfer of technology.
NASA supports an aircraft icing technology program The icing program is guided by three strategic at the NASA Lewis Research Center in Cleveland, Ohio. objectives: Although Lewis Research Center is mainly responsible (1) To develop and validate a system of computer for propulsion and power, it is also responsible for codes that will numerically simulate an aircraft's icing because the NASA Icing Research Tunnel (IRT) is response to an in-flight icing encounter.
located at Lewis. The IRT was built in 1944, during (2) To provide experimental facilities that accu- World War II, and it was used heavily to support the rately simulate the natural icing environment and to development of ice protection systems for the military develop new experimental capabilities and techniques to aircraft of that era. The tunnel has been in operation help the user-community fully utilize these facilities.
ever since 1944, but its use was rather limited for the (3) To support the development and evaluation of period between 1955 and 1978, during which time there advanced ice protection concepts that offer alterna- was no formal icing research program at Lewis. tives to compressor bleed air and other energy- In 1878, because of the strong need expressed by intensive anti-icing systems.
both the U.S. and the European aircraft icing communi- The aircraft industry has emphasized four key pay- ties, NASA reestablished its icing program at Lewis. offs from the NASA icing program: Since 1978, use of the IRT (by NASA, the military, and (1) Validated computer codes and accurate experi- industry) has increased steadily to the point where it mental icing simulations will substantially reduce has become one of NASA's busiest wind tunnels. The developmental and certification testing. This trans- icing technology program at Lewis has likewise lates into reduced time and costs.
increased steadily since 1978. It is interesting to (2) Numerical simulation will reduce the high risk note that Europe also has a strong and growing aircraft of flight testing in icing conditions as these simula- icing technology program.
tions become accepted as an alternative to some flight testing. A complete numerical simulation of an air- The basic codes in use or under development by NASA are: craft's response to an icing encounter appears possible codes combined with water droplet trajec- and economically feasible in the future. (1) Flow (3) Accurate numerical simulations will allow ear- tory codes (2) Ice accretion codes lier assessment of the effect of ice protection requirements on new aircraft designs. This assessment (3) Flow codes for predicting component and is especially important for future military aircraft aircraft performance in icing that require severe weather capability and low observ- (4) Thermal deicer codes These codes are being used, modified, or developed to ability, where ice protection must be considered in the conduct icing simulations on fixed-wing lifting sur- initial design stages.
faces, engine inlets, and rotor blades. Industry uses (4) Advanced, low-power deicers now under develop- ment may offer viable alternatives to conventional these codes (or modifies them and incorporates them bleed systems or energy-intensive electrothermal anti- into their own codes) for preliminary design studies, aero-performance predictions, design and analysis of icing systems. Next generation aircraft will be pow- proposed ice protection systems, and analytical support ered by advanced turbofan engines with higher bypass ratios and smaller core flows. Since the first prior- of the icing certification or qualification process.
ity for compressor bleed is cabin pressurization and Later in our program, codes from the above list air conditioning, there may be inadequate bleed for will be integrated into a system of codes that will conventional hot air anti-icing.
numerically simulate the response of a complete air- Besides receiving support from NASA, the program craft to an in-flight icing encounter.
is also supported by the FAA Technical Center, the Pro- pulsion Directorate of the U.S. Army Aviation Systems Flow/droplet trajectory codes. These codes were Command, and the U.S. Air Force Air Logistics Command. discussed in Refs. 1 to 3, and such codes are part of We also participate in joint or cooperative programs the ice accretion codes, which are discussed below.
with industry, the FAA, the DOD, and academia. These cooperative efforts avoid duplication of resources and Ice accretion codes. Figure 2 shows predicted ice facilities and expedite technical communication and growths on an airfoil at sequential times during expo- transfer of technology to the user community. sure to an icing cloud. These shapes were predicted with the NASA LEWICE two-dimensional (2D) ice accretion COMPUTER ICING SIMULATION prediction code (Ref. 6). The flow field streamlines calculated with a 2D Navier-Stokes solver are also As noted in the INTRODUCTION, a key strategic shown in Fig. 2. Notice the separation bubble and objective of the icing program is to develop and vali- reattachment zone behind the ice shape. LEWICE predic- date a system of computer codes that will numerically tions agree well with ice formed on airfoils during simulate an aircraft's response to an in-flight icing icing tests in the IRT and also with ice formed on the encounter. Selected examples of our code development Twin Otter Icing Research Aircraft during flights in work and code validation experiments are presented in naturally occurring supercooled clouds (Ref. 7). The this section. LEWICE code has been distributed to over 35 organiza- tions, and we continue to receive about one request per Computer Codes and Validation month.
The key phenomena that must be adequately modeled In our continuing efforts to upgrade and enhance in any airfoil icing analysis are illustrated on the the capabilities of LEWICE, we support fundamental left in Fig. 1. These include the flow field around an studies on the physics of the ice accretion process airfoil with leading edge ice, which can cause flow (Refs. 8 to 11). Figure 3 shows a test setup used to separation and reattachment; water droplet impingement observe ice growth on a cylinder. By illuminating the locations and flux; surface roughness; heat and mass surface of the ice with a laser sheet, a time history transfer; and the thermodynamic energy balance that of the ice profile was constructed (Ref. 11). This determines where ice forms. sequence of profiles suggested a multizone heat trans- fer model that was different from that used in LEWICE.
Icing tunnel tests and flight tests in natural icing conditions provide actual ice shapes and aero- At the right of Fig. 3 is the ice shape predicted performance data that can be used to validate code by a version of LEWICE modified to include the multi- zone heat transfer model (Ref. 11). The analysis and predictions. An airfoil under test in the NASA Icing Research Tunnel is shown on the right in Fig. 1. experiment agree remarkably well. This multizone model is undergoing further study and refinement, especially regarding surface roughness and its effect on heat transfer and transition location.
Flow re-attachment
zone
Separated flow
NACA 0012 airfoil, 0° AOA
Streamlines N-S flow field prediction ------------ ► Droplet trajectories Surface roughness Ice accretion
LEWICE ice shape prediction
Figure 2.—Ice accretion prediction with LEWICE and flowfield prediction with ARC21D.
Figure 1.—Ice accretion modeling and experimental validation.
.07
• Experimental
O Computational
.06 •
.05 0
c
` 0 .04
a^
U .03
Hipna = v- • O
o .02
p O
O
.01
Particles trajectories for sweep angle of 30' 0 1 1 1 1 1 1 Figure 6.—Navier-Stokes predictions of streamlines about a swept wing with leading edge ice.
-30 -25 -20 -15 -10 -5 Total temperature, 'C Figure 3.—Ice shapes and drag coefficients; pre- 2D flow codes for predicting performance. Ice on dicted versus experimental.
an airfoil causes decreased maximum lift, decreased stall angle, increased stall speed, and increased drag.
Thus, we need to predict not only the ice shape, but also its effect on aeroperformance. We have therefore upgraded the LEWICE code by incorporating an interac- tive boundary layer (IBL) code that calculates lift and drag changes (Refs. 12 and 13). Figure 4 compares the LEWICE-IBL predictions (Ref. 13) for ice shape and drag Laser with corresponding experimental results (Ref. 8). Ice o Glass rod was accreted on a 21-in. chord, NACA 0012 airfoil in the IRT. Air temperature has a strong effect on ice Experimental results shape and its resultant drag. Runs were made at sever- Test • ; al different air temperatures, while cloud conditions cylinder J and airspeed were kept the same. The predicted and Tunnel floor Multlzone LEWICE measured ice shapes agreed well, as did the drag coef- predictlon ficients. Especially encouraging was the ability to Laser sheet predict the dramatic drag increases observed experimen- tally at temperatures near -5 'C.
3D flow codes. The aerodynamics of modern swept wing aircraft is dominated by three-dimensional effects. NASA therefore supports development of 3D flow codes that can model the flow over swept, semi- span wings with leading edge ice. Figure 5 shows the Figure 4.—Incorporation of new code routines into LEWICE to geometry and gridding used in a Navier-Stokes analysis more closely model observed physics. of a 30' sweep, semi-span wing attached to a vertical wall (Refs. 14 to 16). The leading edge coordinates include a leading edge ice shape. This particular air- foil geometry has the same coordinate geometry as an actual model airfoil being tested in a dry-air wind tunnel. The wind tunnel testing is designed to provide a comprehensive aerodynamic data base for validating 3D viscous codes (Refs. 17 and 18).
Results from the Navier-Stokes analysis (Ref. 15) are shown in Fig. 6 as streamlines above the surface of the swept wing. At 4' angle of attack, a small separa- tion bubble exists behind the ice. Near the leading edge, the separation bubble vortex has a strong span- wise component that grows larger as it moves outboard.
But for the most part, the flow reattaches and we should not expect to see large losses in lift.
At 8' angle of attack, the leading edge ice causes a larger separation bubble. The resulting leading edge vortices have a strong spanwise component that grows very large as it moves outboard. Much of the outboard section of the wing is in separated or reverse flow, so we should expect to see a large dropoff in lift as we Figure 5.—Grid for Navier-Stokes analysis of a swept, finite-length move outboard. It is interesting to note that the ice wing and vertical tunnel wall.
causes a leading edge stall, as opposed to the more familiar bluff body stall that starts at the airfoil trailing edge and moves further forward with increased angle of attack.
Although the Navier-Stokes analysis is very accu- rate, it requires long run times on a supercomputer.
An interactive boundary layer (IBL) code coupled with a potential flow code models less of the detailed phys- ics, but requires far less computational time. At this time, NASA is supporting development of both approaches because the IBL approach has potential to be a good engineering design tool, and the Navier-Stokes approach has the potential for accurately modeling the detailed physics.
Figure 7 shows a 30° swept, semi-span wing model a dry air with simulated leading edge ice installed in wind tunnel (Refs. 17 and 18). The wind tunnel testing is designed to provide a comprehensive aerodynamic data Figure 7.—Swept, finite-length wing with leading edge ice shape base for validating 3D codes, such as the Navier-Stokes installed in dry air wind tunnel.
analysis and the 3D interactive boundary layer (IBL) analysis.
This model has five chordwise rows of surface static pressure taps, and is attached to a three- component force balance in the wind tunnel wall. Flow diagnostics include laser velocimetry, laser sheets, 1.0 and helium bubble seeding and tracking.
Experiment El Figure 8 shows a comparison between the Navier- Calculation Stokes predictions discussed above and the wind tunnel results for lift coefficient versus span, at 4° and 8° c angles of attack (a) (Ref. 15). The agreement is C)
nearly perfect. U .6
In the dry-air wind tunnel testing of this model a=8deg
(Refs. 17 and 18), the flow was seeded with helium bub- o
L)
r .4
bles and high-speed videography was used to observe the bubbles' trajectories. The experimentally observed
J
helium bubble trajectories will be compared with the .2 ^_ a_4 deg __0 streamlines predicted by the Navier-Stokes code.
Airplane performance in icing. A major result from the numerical simulation of an aircraft icing encounter .2 .4 .6 .8 1.0 is the predicted changes in aircraft performance and Span stability caused by ice. Thus, NASA is supporting Figure 8.—Lift coefficient along span of iced wing with development of a computer code that will predict per- formance and stability of modern aircraft with given 30 0 leading edge sweep; comparison between ice shapes on the lifting surfaces. This work will be Navier-Stokes predictions and wind tunnel data.
carried out along with ongoing efforts to develop ice accretion codes for 3D surfaces. Later, at the appro- priate time, the ice accretion codes will be incorpo- rated into the aircraft performance and stability code.
Figure 9 shows the pressure distribution over the NASA Twin Otter icing research aircraft as calculated with a nonviscous panel code. Flight data from the Surface pressure coefficients over NASA Twin Otter will be used to validate the code at full- icing research aircraft scale Reynolds numbers. First the Twin Otter will be flown in clear air with 'styrofoam • ice shapes on its tail surfaces (Refs. 19 and 20), and next it will be flown in naturally occurring supercooled clouds.
In addition to flight testing with the Twin Otter, NASA will conduct dry-air wind tunnel testing of a sub- scale model of a modern swept-wing aircraft with simu- lated ice on its lifting surfaces. The wind tunnel results will provide code validation data for a modern aircraft configuration. This will be a joint program between NASA Lewis and NASA Langley. After a good experimental data base has been acquired for a subscale model in dry air wind tunnel, we plan to conduct flight testing with a modern swept wing aircraft to acquire a validation data base at full-scale Reynolds numbers.
Figure 9.—VSAERO predictions for surface pressure coeffi- Thermal deicer codes. NASA has sponsored the development of a series of transient heat conduction cients over NASA Twin Otter icing research aircraft.
codes that numerically model electrothermal deicer NASA Icing Research Tunnel operation. These codes, developed by the University of Icing wind tunnels undoubtedly offer the most Toledo, are used by industry to analyze and design versatile approaches to icing testing. It generally thermal deicer systems. The model consists of electri- costs much less to test components in an icing wind cal heater strips surrounded on one side by insulation tunnel than in flight, and conditions can be much more and aircraft structure, and on the other side by insu- closely controlled and repeated. In icing tunnel test- lation and the wing leading edge covered with ice. The ing, productivity is high, and safety risk is very low.
codes calculate the temperature distribution inside the But there definitely is an appropriate role for flight deicer-wing-ice assembly and determine the heat testing, as discussed elsewhere (Refs. 3, 5, 19, required to melt the ice at the ice-wing interface.
The codes also include the ability to follow the melt and 20).
A schematic of the NASA Lewis Icing Research Tun- line which has water on one side and ice on the other.
Reference 21 reviews the codes developed by the Univer- nel is shown in Fig. 10. The components shown in the sity of Toledo up to 1988. Recently, Toledo has incor- inserts are upgraded systems that were installed when porated an electrothermal deicer analysis capability the IRT was rehabilitated in 1986-87. In addition to having all the systems of a conventional dry-air tun- into LEWICE (Ref. 22). This was accomplished by replacing a subroutine in LEWICE that balanced the nel, an icing tunnel has two unique systems: (1) a water spray system that injects water droplets into the energies at the ice surface, with a subroutine that performs this same energy balance, as well as calcu- airstream to create a supercooled cloud and (2) a lates all the time-temperature transients below the ice refrigeration system and heat exchanger that cools the air to temperatures as low as -20 °F. The heat surface, for all the layers of the deicer and wing as well as within the ice layer itself. This enhancement exchanger is in the leg just upstream of the spray to LEWICE allows us to calculate the dynamic processes bars. Closed-loop refrigerated tunnels can "dial in of ice growth, ice melting, and ice shedding. This new the weather' any time of the year and are therefore capability should prove useful for determining optimum very productive.
heater power levels and heater on-off times required to The IRT is the largest refrigerated icing wind tunnel in the world. The test section is 6 ft high by melt and shed ice with minimum power usage and with avoidance of water runback and freeze beyond the 9 ft wide by 20 ft long. The maximum airspeed for an deicers. empty IRT test section is 300 mph, but model blockage greater than 20 percent significantly reduces maximum EXPERIMENTAL ICING SIMULATION airspeed. The nozzle spray system produces supercooled clouds that can be controlled over a range of liquid and water droplet NASA has two major commitments in our strategic water contents from 0.5 to 2.5 g/m goal for experimental icing simulation: first, to median volumetric diameters from 15 to 40 jam.
provide experimental facilities that accurately simu- The IRT is one of NASA's busiest wind tunnels; in 1988 it logged 1330 hr of actual test time. It carries late the natural icing environment; and second, to a 2-year backlog of research and development testing develop new experimental capabilities and test tech- niques to help the user community fully utilize these for NASA, the military, and industry. We conduct many facilities. Examples of these new capabilities and joint programs with the military and with industry.
techniques are the subscale rotor testing in the NASA Reports describing the IRT and its calibration are Icing Research Tunnel (IRT), which will be discussed given in Refs. 23 to 25.
In 1987, the American Society of Mechanical Engi- below, and a new three-component force balance for the neers (ASME) designated the Lewis Research Center's IRT IRT.
an International Historic Mechanical Engineering Land- The NASA Icing Research Tunnel and the NASA Twin mark for its leading role in making aviation safer for Otter flight research aircraft are used extensively for everyone (Fig. 11).
code validation, advanced ice protection development, and in the case of the IRT, for testing actual aircraft components.
"VARICHRON SYSTEM" "SPRAY BAR CONTROLS BUILDING" "6 FT x9 FT 20 FT LONG" Figure 11.—Photo of plaque that designates the IRT an ASME "SPRAY BAR SYSTEM" "CONTROL ROOM" International Historic Engineering Landmark.
Figure 10.—NASA Lewis Icing Research Tunnel.
NASA Twin Otter Flight Research Aircraft Flight testing in natural icing is currently the The NASA Lewis icing research aircraft shown in only acceptable means for certifying that a helicopter Fig. 12 is a modified DeHavilland DH-6 Twin Otter rotor can perform safely in the icing environment (Refs. 26 to 28). The aircraft maximum range for icing defined by the full Federal Aviation Administration research flights is 300 nmi. Cruising speed is 170 kn (FAA) FAR Part 25 Appendix C icing envelopes. The U.S.
at sea level and 182 kn at 10 000 ft. An oxygen system Rotorcraft industry estimates that it would cost about is available for altitudes up to 15 000 ft. 15 million dollars to certify a helicopter in natural The aircraft is equipped with electrothermal anti- icing to the FAR 25 requirements; and they feel that icers on the propellers, engine inlets, and windshield. this cost is prohibitively high. Although it is not well-known, only one civilian helicopter is certified Pneumatic deicer boots are located on the wing outboard by the FAA for flight into known icing conditions; that of the engine nacelles, on both the horizontal and vertical stabilizers, on the wing struts, and on the helicopter is the French Super Puma. it took the rear landing gear struts. The pneumatic deicers French nearly 10 years of flight testing in natural icing to win that certification.
located on the vertical stabilizers, wing struts, and For several years, NASA and the U.S. rotorcraft landing gear struts are nonstandard items that provide industry have been engaged in a joint effort to develop additional research capability for measuring component new methods that could help reduce the cost and time drag through selective deicing. The aircraft is needed to certify and qualify U.S. rotorcraft for equipped with several standard instruments for measur- icing. These methods include (1) computer codes that ing icing cloud properties (Ref. 29).
reliably predict full-scale rotor performance in icing Wing leading edge ice shapes are measured in and (2) experimental techniques for testing subscale flight with a stereo photography system. Wing section model helicopter rotors in the IRT to acquire data for drag is measured with a wake survey probe mounted on validating the codes and to develop a better under- the wing behind the region where the stereo photos are standing of the effects of icing on rotor performance.
taken. A noseboom is used to measure airspeed, angle- The methods derived from this joint effort will also of-attack, and sideslip. A flight test system measures advance the state-of-the-art methods for predicting the flight dynamics along a flight path. The system effects of ice accretion and shedding for the Advanced includes a data acquisition system and an inertial Ducted Propeller and other thrusting devices.
package that contains rate gyros, directional gyros, Figure 13 shows a subscale helicopter being tested and servo accelerometers.
in the IRT. The model consists of a UH-60 Blackhawk The icing flight research aircraft acquires helicopter fuselage, four NACA 0012 blades (5-in.
in-flight data that can be used to validate ice accre- chord, 6-ft diam), a fully articulated rotor head, and tion and aeroperformance computer codes and to confirm a six-component force balance housed under the Black- that the IRT adequately simulates natural icing. As hawk fuselage.
noted above, NASA is developing an airplane performance Figure 14 shows the rotor torque rise caused by and stability code for modeling flights in icing. The ice accretion versus time in icing for the model shown Twin Otter, with its flight test package, is being used in Fig. 13. The experimental results are compared with to acquire a performance and stability data set for an analytical prediction developed by Flemming calibration and validation of these codes.
(Refs. 30 and 31). The rotor icing analysis includes an ice shedding model, which is necessary for good Subscale Rotor Testing in the IRT agreement with the experiment. The analysis also This section on subscale rotor testing in the IRT includes empirical airfoil performance-in-icing data illustrates a recent example of NASA's commitment to that was acquired in another test program funded by develop new test techniques to help the user community NASA (Ref. 30). The comparison between theory and fully utilize the IRT.
experiment, as shown here, is remarkably good for the conditions of this particular test run. Similar agree- ment between theory and experiment was also found for lift loss versus time in icing.
STEREOSCOPHIC PHOTOGRAPHY HEATED WAKE f-tn SURVEY PROD[_ -I SYSTEM, \ 1505NA • _ ` V
Au
Scale model of UH-60 Blackhawk helicopter with 4 NACA 0012 blades, fully articulated rotorhead, and 6-component force balance Figure 13.—Subscale model rotor testing in the IRT.
Figure 12.—NASA Lewis Icing Research Aircraft.
pattern) embedded in elastomeric material that fits over the outside of the wing leading edge. When a capacitor is discharged into either type of spiral conductor, the current pulse causes a rapidly changing magnetic field that induces eddy currents in the air- foil metal leading edge. The eddy currents and spiral currents produce opposing magnetic fields that rapidly m deform the leading edge. Pneumatic impulse deicers
Cr have tubes underneath a boot that covers the leading
L
edge of the wing. A high pressure air pulse inflates ° 40 the tubes and rapidly distorts the boot.
O
Mr. Len Haslim, of NASA Ames Research Center,
O
invented the electro-expulsive deicer system. Data
cc
Products of New England (ONE) has purchased the NASA
a 20
rights to Haslim's invention and are developing the system further (Ref. 35).
Since, by definition, deicers allow ice to accumu- late on the aircraft surfaces before the deicers are actuated, ice particles will shed from the surfaces 40 60 80 20 during actuation. If deicers are used on engine inlets, the engine must ingest ice particles without
Icing time, sec
sustaining damage to fan blades or other components.
For this reason, NASA has initiated the development of Figure 14.—Delta rotor torque versus time in icing.
a structural analysis code for determining the response of engine fan blades to ice impact.
The four U.S. helicopter companies have received USAF/NASA low-Dower ice p rotection technoloav Dro- all the data from this test program. Selected results gram. Our current goal is to develop an experimental from this program, which include both analytical model- data base for the low-power impulse deicers. To that ing and experimental validation, have been reported in several technical reports (Refs. 31 to 34). The com- end, we have conducted a joint USAF/NASA/industry pro- plete results will be published in a final contractor gram to test promising impulse deicers systems in the NASA IRT. In this test program, a total of eight report. We are also planning a second test entry into impulse deicers systems, supplied by six companies, the IRT to expand the original icing flight test enve- were individually tested in the IRT under identical lope and to further study rotor performance at warmer conditions (Ref. 36).
outside air temperatures.
Figure 15 shows an airfoil with a deicer system installed on the leading edge. Although not obvious ADVANCED ICE PROTECTION SYSTEMS from the photo, the deicer boot covers about the first 15 percent of the airfoil chord. This airfoil geometry As mentioned in the INTRODUCTION, NASA also sup- ports the development and evaluation of advanced ice was chosen because its 0.5-in. leading edge radius protection systems that offer lower-power alternatives to evaporative anti-icing systems, such as electrother- mal or compressor bleed air systems. One promising alternative is the new class of electromechanical and pneumatic impulse deicers that use only 10 to 20 per- cent of the power used by evaporative anti-icing sys- tems. Impulse systems have pulse times less than a millisecond and surface accelerations up to 1000 g's, which impart forces strong enough to shatter, debond, and expel the ice. The impulse systems have the poten- tial for maintaining ice thicknesses very thin, both before and after actuation.
In broad terms, impulse systems fall into one of three categories: electro-expulsive, eddy-current, and high pressure pneumatic. The first two approaches employ a capacitor bank energy storage system that sup- plies a short, high pulse of electrical current to pro- duce a repulsive action between two conductors that rapidly distorts the airfoil's leading edge. The third approach uses a short pulse of high pressure gas to achieve the distortion.
Electro-expulsive deicers consist of a double layer of electrically conducting strips in an elasto- meric blanket that covers the leading edge. Current discharges into the top and bottom conductors in oppo- sition, which produces opposing magnetic fields that rapidly force the strips apart. Eddy-current repulsion deicers are divided into two types: The first (known as electromagnetic impulse deicers) employ thin spi- rally wound pancake solenoidal coils that fit inside Figure 15.—NACA 0012 airfoil (21-in. chord) installed in the wing, up against the leading edge. The second IRT. Deicer systems were applied to the leading edge (known as eddy current repulsion deicer) employs a flat of the airfoil.
conductor sheet (in which is cut a spiral conductor represented a challenge for most manufacturers, and because we were trying to simulate the small leading edge radii used on inlets of some military aircraft.
In testing impulse deicers the following parame- ters were measured to characterize deicer performance: (1) maximum size of shed ice particles for a given ice thickness and pulse energy; (2) minimum thickness of ice that can be removed for a given pulse energy; A (3) amount, texture, and height of residual ice remain- ing on the surface before and after deicer actuation for several different times between actuations; (4) energy per unit area or per unit span length required for one deicer actuation; and (5) weight per unit area r of deicer coverage.
In evaluating deicer performance, the systems must be tested under the full range of expected icing condi- tions. Experience has shown that two conditions give impulse deicers the most trouble: near-freezing condi- tions that produce soft, mushy ice with water between Figure 17.—Eddy-current repulsion deicer actuation during devel- the ice and deicer surface; and cold, rime icing condi- opment tests in IRT as part of a NASA SBIR.
tions that cause the ice to adhere strongly to the deicer (Refs. 37 and 38).
Figure 16 shows a sequence of photos that capture FY 91 92 93 94 95 96 98 an ice shedding event by means of high speed videogra- Fixed wing phy. Events can be captured at speeds up to 6000 frames per second. The video tape provides a digitized record that can be examined frame by frame on a conven- Ice Performance Numerical protection & stability simulation of tional computer monitor to follow the ice breakup pro- concepts code overall aircraft cess. The digitized data also allows estimating the In Icing size of the largest particles shed during an actuation.
Analysis codes Special image processing software is being developed to Icing physics Generic research automate the estimation of particle sizes and possibly Instrumentation Ice prevention and to obtain size spectrums as well.
removal concepts Figure 17 shows ice being expelled from a cylinder Rotor Ice Numerical Subscale by an eddy-current repulsion deicer strip that was erformance protection simulation rotor Icing undergoing development testing in the IRT. The work P n Icing concepts main rotor test code In Icing techniques was done under a NASA Small Business Innovative Research Contract. This particular deicer can be ap- plied over the outside of a component; it causes only Rotary wing minimal intrusion into the component.
Figure 18. —NASA Aircraft Icing Technology Program Plan.
AIRCRAFT ICING TECHNOLOGY PLAN The Aircraft Icing Technology Program has a rotor test techniques, and advanced ice protection concepts are continually worked throughout the program strong, focused research effort supporting the strate- and results are promptly delivered to industry for gic objectives for both fixed and rotary wing aircraft (see Fig. 18). The various analytical codes that sup- inclusion in their own icing program.
port ice accretion, aeroperformance, and ice protection are developed in the focused research. Companies and CONCLUDING REMARKS Government agencies receive these codes while they are still in the research stage so that NASA may get feed- The key strategic objectives of NASA's Aircraft back on the user's experience with the codes and on Icing Technology Program are (1) to numerically simu- desired additional capabilities. Icing physics late an aircraft's response to an in-flight icing encounter, (2) to provide improved capabilities and research supports the development of analytical models techniques for ground and flight icing testing, and for ice accretion, ice shedding, and ice removal.
(3) to offer innovative approaches to ice protection.
Droplet sizing instrumentation is essential for vali- dating droplet trajectory codes and ice accretion With a comprehensive computer code development program in place, we are progressing toward producing a codes.
methodology for numerically simulating the response of Figure 18 shows that while the strategic objec- tives are met in the outyears, codes, subscale model a complete aircraft to an icing encounter. At the same time, the codes are being used extensively by industry and Government in support of their icing programs.
Through a strong joint program with the U.S. heli- P copter industry, we have demonstrated that subscale y model rotor testing in an icing wind tunnel provides valuable data for developing and validating computer codes that predict rotor performance in icing. The encouraging progress to data justifies further work in subscale model rotor testing in support of icing i certification.
Figure 16.— High-speed videography to capture ice shedding frames/sec).
event (1000 Shin, J., et al.: Prediction of Ice Shapes and Through our joint USAF/NASA/industry test program 13.
Their Effect on Airfoil Performance. AIAA Paper we have succeeded in developing an extensive, but pre- (Also, NASA TM-103701.)
91-0264, Jan. 1991.
liminary, data base on the new class of electromechani- cal and pneumatic impulse deicers. Because each 11. Kwon, O.J.; and Sankar, L.N.: Numerical Study of impulse deicer needs a detailed evaluation under a wide the Effects of Icing on Finite Wing Aerodynamics.
range of icing conditions and under various operating AIAA Paper 90-0757, Jan. 1990.
modes, these systems will require much more testing.
Our good working relationships with industry, Kwon, O.J.; and Sankar, L.N.: Numerical Study of academia, and other Government agencies results in a 15.
combination of our individual resources, avoids dupli- the Effects of Icing on Fixed and Rotary Wing cation of effort and facilities, and expedites technol- Performance. AIAA Paper 91-0662, Jan. 1991.
ogy transfer to the user community.
Potapczuk, M.G., et al.: Simulation of Iced Wing 16.
Aerodynamics. NASA TM-104362, 1991.
REFERENCES 1. Reinmann, J.J.; Shaw, R.J.; and Ranaudo, R.J.: 17. Bragg, M.B.; and Khodadoust, A.: Effect of Simu- NASA's Program on Icing Research and Technology.
lated Glaze Ice on a Rectangular Wing. AIAA Paper Flight in Adverse Environmental Conditions, AGARD 89-0750, Jan. 1989.
CP-470, 1989, pp. 22-1 to 22-31. (Also, NASA TM-101989.)
18. Bragg, M., et al.: Effect of a Simulated ice Ac- cretion on the Aerodynamics of a Swept Wing. AIAA 2. Shaw, R.J.; Potapczuk, M.G.; and Bidwell, C.S.: Paper 91-0442, Jan. 1991.
Predictions of Airfoil Aerodynamic Performance Degradation Due to Icing. Fourth Symposium on 19. Ranaudo, R.J., et al.: Determination of Longitudi- Numerical and Physical Aspects of Aerodynamic nal Aerodynamic Derivatives Using Flight Data from Flows, California State University, 1989, NASA an Icing Research Aircraft. AIAA Paper 89-0754, TM-101434.
Jan. 1989. (Also, NASA TM-101427.)
3. Ranaudo, R.J.; Reehorst, A.L.; and Potapczuk, M.G.: 20. Batterson, J.G.; and O'Mara, T.M.: Estimation of An Overview of the Current NASA Program on Air- Longitudinal Stability and Control Derivatives for craft Icing Research. SAE Paper 881386, Oct.
an Icing Research Aircraft from Flight Data. NASA 1988.
TM-4099, 1989.
4. Reinmann, J.J.: Selected Bibliography of NACA-NASA 21. Keith, T.G., et al.: Overview of Numerical Codes Aircraft Icing Publications. NASA TM-81651, 1981.
Developed for Predicted Electrothermal De-Icing of Aircraft Blades. AIAA Paper 88-0288, Jan. 1988.
5. Potapczuk, M.G.; and Reinmann, J.J.: Icing Simula- tion, a Survey of Computer Models and Experimental 22. Wright, W.B.; Keith, T.G.; and DeWitt, K.J.: Facilities. NASA TM-104366, 1991. Numerical Simulation of Icing, Deicing, and Shed- ding. AIAA Paper 91-0665, Jan. 1991.
6.
Ruff, G.A.; and Berkowitz, B.M.: Users Manual for the NASA Lewis Ice Accretion Prediction Code 23. Soeder, R.H.; and Andracchio, C.R.: NASA Lewis (LEWICE). NASA CR-185129, 1990.
Icing Research Tunnel User Manual. NASA TM-102319, 1990.
7. Berkowitz, B.M.; and Riley, J.T.: Analytical Ice Shape Predictions for Flight in Natural Icing 24. Ide, R.F.: Liquid Water Content and Droplet Size Conditions. NASA CR-182234, 1988.
Calibration of the NASA Lewis Icing Research Tun- nel. AIAA Paper 90-0669, Jan. 1990. (Also, NASA 8. Olsen, W.A.; Shaw, R.J.; and Newton, J.: Ice TM-102447, AVSCOM TM 89-C-014.)
Shapes and the Resulting Drag Increase for a NACA 0012 Airfoil. AIAA Paper 84-0109, 1984. (Also, 25. Newton, J.E.; and Olsen, W.A.: Study of Ice Accre- NASA TM-83556.) tion on Icing Wind Tunnel Components. AIAA Paper 86-0290, Jan. 1986. (Also, NASA TM-87095.)
9. Olsen, W.; and Walker, E.: Experimental Evidence for Modifying the Current Physical Model for Ice 26. Mikkelsen, K.L., et al.: Icing Flight Research: Accretion on Aircraft Surfaces. NASA TM-87184, Aerodynamic Effects of Ice and Ice Shape Documen- 1986.
tation With Stereo Photography. AIAA Paper 85-0468, Jan. 1985. (Also, NASA TM-86906.)
10. Hansman, R.J., et al.: Modeling of Surface Rough- ness Effects on Glaze Ice Accretion. J. 27. Mikkelsen, K., et al.: In-flight Measurements of Thermophys. Heat Trans., vol. 5, no. 1, Jan. 1991, Wing Ice Shapes and Wing Section Drag Increases pp. 54-60.
Caused by Natural Icing Conditions. NASA TM-87301, 1986.
11.
Yamaguchi, K.; Hansman, R.J.; and Kazmierczak, M.: Deterministic Multi-Zone Ice Accretion Modeling. 28.
Ranaudo, R.J., et al.: The Measurement of Aircraft AIAA Paper 91-0265, Jan. 1991.
Performance and Stability and Control After Flight Through Natural Icing Conditions. AIAA Paper 12. Cebeci, T., et al.: Prediction of Post-Stall Flows 86-9758, Apr. 1986. (Also, NASA TM-87265.)
on Airfoils. Fourth Symposium on Numerical and Physical Aspects of Aerodynamic Flows, California State University, 1989.
29. Ide, R.F.; and Richter, G.P.: Comparison of Icing 34. Britton, R.; and Bond, T.H.: A Review of Ice Accretion Data From a Model Rotor Icing Test and Cloud Instruments for 1982-1983 Icing Season Flight Program. AIAA Paper 84-0020, Jan. 1984. Comparison to Theory. AIAA Paper 91-0661, Jan.
(Also, NASA TM-83569.) 1991. (Also, NASA TM-103712.)
30. Flemming, R.L.; and Lednicer, D.A.: High Speed Ice 35. Goldberg, J.; and Lardiere, B.: Developments in Accretion on Rotorcraft Airfoils. NASA CR-3910, Expulsive Separation Ice Protection Blankets.
1985. AIAA Paper 89-0774, Jan. 1989.
Flemming, R.J.; Bond, T.H.; and Britton, R.K.: 36. Bond, T.H.; Shin, J.; and Mesander, G.A.: Advanced 31.
Model Rotor Icing Tests in the NASA Lewis Icing Ice Protection Systems Test in the NASA Lewis Icing Research Tunnel. NASA TM-103757, 1991.
Research Tunnel. NASA TM-104351, 1991.
Bond, T.H.; Flemming, R.J.; and Britton, R.K.: 37. Zumwalt, G.W., et al.: Electro-Impulse De-Icing 32.
Testing Analysis and Design. NASA CR-4175, 1988.
Icing Tests of a Sub-Scale Model Rotor. Proceed- ings of the 46th Annual American Helicopter Soci- ety Forum, May 1990, pp. 267-281. 38. Martin, C.; and Putt, J.: An Advanced Pneumatic Impulse Ice Protection System (PIIP) for Aircraft.
33. AIAA Paper 90-0492, Jan. 1990.
Flemming, R.J.; Bond, T.H.; and Britton, R.K.: Results of a Sub-Scale Model Rotor Icing Test.
AIAA Paper 91-0660, Jan. 1991. (Also, NASA TM-103709.)
Nan
Report Documentation Page
National Aeronautics and Space Administration 2. Government Accession No. 3. Recipient's Catalog No.
1. Report No.
NASA TM-104518
5. Report Date
4. Title and Subtitle
NASA's Aircraft Icing Technology Program
6. Performing Organization Code 8. Performing Organization Report No 7. Author(s)
John J. Reinmann
E-6388
10. Work Unit No.
505-68-11
9. Performing Organization Name and Address
11. Contract or Grant No.
National Aeronautics and Space Administration
Lewis Research Center
Cleveland, Ohio 44135 - 3191
13. Type of Report and Period Covered
Technical Memorandum
12. Sponsoring Agency Name and Address
National Aeronautics and Space Administration
14. Sponsoring Agency Code
Washington, D.C. 20546-0001
15. Supplementary Notes
^I
Prepared for the 1991 Winter Annual Meeting of the American Society of Mechanical Engineers, Atlanta, Georgia, December 1-6, 1991. Responsible person, John J. Reinmann, (216) 433-3900.
16, Abstract
NASA's Aircraft Icing Technology program is aimed at developing innovative technologies for safe and efficient flight into forecasted icing. The program addresses the needs of all aircraft classes and supports both commercial and military applications. The program is guided by three key strategic objectives: (1) numerically simulate an aircraft's response to an in-flight icing encounter, (2) provide improved experimental icing simulation facilities and testing techniques, and (3) offer innovative approaches to ice protection. Our research focuses on topics that directly support stated industry needs, and we work closely with industry to assure a rapid and smooth transfer of technology. This paper presents selected results that illustrate progress toward the three strategic objectives, and it provides a comprehensive list of references on the NASA icing program.
17. Key Words (Suggested by Author(s)) 18. Distribution Statement
Ice formation; Aerodynamic characteristics; Unclassified - Unlimited
Aircraft performance; Ice prevention; Cold
Subject Categories 01 and 02
weather tests
19. Security Classif. (of the report)
20. Security Classif. (of this page) f pages 22. Price' 21. No. o
Unclassified Unclassified
A03
I '
NASA FORM 1626 OCT 86 'For sale by the National Technical Information Service, Springfield, Virginia 22161