Document
N88-16641
NASA's ROTORCRAFT ICING RESEARCH PROGRAM Robert J. Shaw and John J. Relnmann National Aeronautics and Space Administration Lewis Research Center Thomas L. Miller Sverdrup Technology, Incorporated NASA Lewis Research Center SUMMARY The objective of the NASA aircraft icing research program is to develop and make available icing technology to support the needs and requirements of industry for all weather aircraft designs. While a majority of the technol- ogy being developed is viewed to be generic (i.e. appropriate to all vehicle classes), vehicle specific emphasis is being placed on the helicopter due to its unique icing problems. In particular, some of the considerations for O_J rotorcraft icing are indicated in figure I. The NASA icing research program (-,-) & emphaslzes technology development in two key areas: ice protection concepts I_.I and icing simulation (analytical and experimental). The NASA research efforts related to rotorcraft icing in these two technology areas will be reviewed in this paper.
ICE PROTECTION CONCEPTS Currently, the only rotor ice protection system being employed on opera- tlonal military and civilian helicopters is the electrothermal deicing system. While the electrothermal concept has been shown to be capable of effectively deicing rotors, the weight and power requirements have caused the industry to seek other concepts. NASA research efforts have focused on two concepts: pneumatic boot and electro-lmpulse deicers.
PNEUMATIC BOOT The pneumatic boot is probably one of the oldest of all ice protection concepts with its use on fixed wing aircraft dating back to at least the 1930's. The application of the pneumatic boot concept to the helicopter rotor was studied by Lockheed under an Army contract in the 1970's (ref. l) and rejected, primarily on materials concerns. Specifically, it was felt that the pneumatic boot would not withstand the service dynamic environment of the helicopter rotor. It was feared that the boots might be damaged or completely torn off by the high levels of centrifugal forces. Also, it was felt that the raln/eroslon characteristics of the neoprene rubber, from which the boots would be constructed, would be unacceptable. Concern was also expressed about the possible aerodynamic performance degradation of the rotor due to the presence of the boots especially when the tubes were inflated. These concerns were sufficient to eliminate the pneumatic boot from further consideration at that tlme.
However, the B.F. Goodrich Company, the only company marketing pneumatic boots tn thls country, subsequently Investigated other candidate materials for boot construction and became convinced that a polyurethane material (trade name Estane) would solve several of the aforementioned problems.
Furthermore, the possible simplicity, light weight, low power consumption and low cost of a rotor pneumatic boot detclng system Justified pursulng a technology development program. Based on these considerations, NASA and B.F. Goodrtch conducted a Jotnt program In 1979 to Investigate the deicing capability and aerodynamic performance of candidate pneumatic boot designs.
The results of that test program conducted in the NASA Icing Research Tunnel (IRT) are given In reference 2.
The wind tunnel model used In the test (fig. 2) was a 6 ft span segment of a full scale UH1H rotor blade. Of the three pneumatic boot configurations tested, the one Judged to be the most effective at removing Ice Is shown In ftgure 3. Thts configuration had two spanwlse tubes surrounding the leading edge and chordwtse tubes aft of the leadtng edge tubes on both the suctton and pressure surfaces. .- Oelctng performance of this configuration for representative rlme and glaze lclng conditions Is shown In flgure 4. The drag levels as measured by a translating wake survey probe Indicated large drag Increases relative to clean model levels due to the Ice accretions (80 to 171 percent) and drag Increases of 21 to 52 percent due to the residual Ice left after one boot Inflation.
The results of this preliminary, proof-of-concept test were encouraging enough that a Joint NASA/Army/Bell/B.F. Goodrtch program was established to conduct the requlred flight tests to further evaluate the feasibility of the pneumatic boot concept for rotor detclng. This flight testing effort has been conducted In several phases: fltght loads survey performance and handltng qualltles tests, hover tctng (artificial), forward fllght Icing (artificial and natural), and rain and sand erosion testing.
Clear air and tclng fllght tests were conducted by the Army Engineering Fllght Actlvlty (Edwards Air Force Base) while Bell Helicopter Textron Instr- umented the set of rotor blades to which had been afflxed the pneumatlc boot delcers provlded by B.Fo Goodrich. The system as Installed on the JUH-IH welghed only 30 lb and consisted of only six major components: pnuematlc delcer, regulator-reliever shut-off valve, timer, rotary union, hose and flap assembly, and ejector flow control valve to keep the deicers deflated.
Flgure 5 shows a schematic of the JUH-1H hellcopter wtth the key pneumatic deicer components Indicated.
Hover lctng tests were conducted tn the Canadian Nattonal Research Council's Ottawa Spray Rtg (flg. 6) while the forward flight Icing tests were conducted behind the Army's Helicopter Icing Spray System (HISS) tanker (fig. 7). Natural tclng flight tests were conducted In the general vlclnlty of Duluth, RN. Sand and ratn erosion tests were conducted at Fort Rucker, Alabama by the Army Avlatton Development Test Actlvlty (USAAVNDTA). Limited artificial rain eroston fltght tests were conducted at Edwards Air Force Base, CA by the Army Avlatlon Engineering Flight Actlvlty (USAAEFA).
lhe deicing capability of the deicer configurations tested on the JUH-IH was Judged to be satisfactory for the range of light thru moderate icing conditions tested. Structual loads measured during flight tests were all within acceptable endurance limits except the maln rotor pitch llnk axial load which exceeded the limit by 20 percent. This was not Judged to be critical. No unusual dynamic responses were noted and handling qualities of the JUH-IH were essentially unchanged by installation of the pneumatic boots.
However, significant hover and level flight performance degradation due to pneumatic boot installation effects of the JUH-IH was measured. A summary of the flight test efforts Is given in reference 3.
As already indicated, the ability of the Estane material to withstand the sand and rain erosion environment of the rotor was a key issue regarding feasibility of the pneumatic boot deicer. Ten hours of sand erosion testing at Fort Rucker and 6 hours of artificial raln erosion testing behind the HISS tanker failed to result In any detectable erosion. However, lO hours of flying in natural raln of varying intensity required four separate blade repairs due to pitting and chunking of the Estane material. These repairs were required in the outer span region of the blades. This region was covered by a layer of Estane and was not protected by an active pneumatic deicer, but rather relied on self shedding for deicing.
lhe preliminary results from the NASA/Army/Industry program suggest that the pneumatic boot deicer is a feasible alternate rotor ice protection system. However, additional field testing appears to be needed to galn more experience with the erosion problems including the practicality of making repairs In the field before final conclusions can be drawn.
ELECIROMAGNETIC IMPULSE DEICER The electromagnetic impulse deicer (EIDI) is a potentially attractive system for low weight, low power consumption, highly reliable and efficient deicing of aircraft components. The EIDI concept, while first suggested as early as 1937, had not been adequately developed to allow Industry to con- sider incorporation in current or future aircraft designs. However, NASA sponsored an intensive 4 year Joint effort with industrial and university partners to develop the required technology base for the EIDI concept. Thls activity was completed in 1986. Reference 4 summarizes the NASA industry university program for EIDI technology development.
The basic principles behind the operation of the EIDI system are shown in figure 8. Flat-wound coils made of copper ribbon wlre are placed Just inside the leading edge of a wing's skln wlth a small gap separating skin and coil. Either one or two coils are placed at a given spanwise station, depending on the slze of the leading edge. Two methods of supporting coils are shown: support by the front spar or from a beam attached to the ribs.
Also, mounting to the skin itself is sometimes used.
The coils are connected by low resistance, low inductance cables to a hlgh voltage capacitor bank, and energy is discharged through the coll by a remote signal to a slllcon-controlled-rectlfler ("thyrlstor"). Discharge of the capacitor through the coils creates a rapidly forming and collapsing electromagnetic field which induces eddy currents in the metal skin. The fields resulting from current flow in the coil and skin create a repulsive force of several hundred pounds magnitude, but for a duration of only a fraction of a millisecond. A resultlng small amplitude, high acceleration movement of the skin acts to shatter, debond and expel the ice. Two or three such "hits" are performed sequentially, separated by the time required to recharge the capacitors, then ice is permitted to accumulate until it again approaches an undeslrable thickness.
The EIDI system is lightweight and uses a small amount of electrical energy (about 200 3 per foot). For fixed-wlng alrcaft, system weights are comparable to those for pneumatic deicers, and the power used is about equal to the landing light power. The power required is about one percent of that used by a hot gas antl-lclng system, or about lO percent of an electrothermal system.
The ability of the EIDI system to efficiently deice the many fixed wing icing components which were tested in the IRT suggested that the system might also be used to deice helicopter rotors. In order to provide a preliminary evaluation of this hypothesis, a mldspan section of an AHI Cobra rotor blade was reworked to include EIDI coils and the resulting model was tested in the IRT (fig. 9). The Cobra blade section was of all composite design and had a chord of 31 in., but for these tests the leading edge was made of sheet metal, and attached only at its upper and lower surface extremetles.
Acceptable deicing characteristics (as shown in fig. lO) were observed for a configuration which had mlnl-coils located every 8 inches on the pres- sure surface. Coils placed adjacent to the upper or suction surface did not yield such satisfactory deicing results either when tested alone or in con- Junction wlth selected lower surface coils.
While these initial fixed position airfoil de-iclng tests gave satisfac- tory deicing results, several design related problems became apparent. In particular, the rotor leading edge abrasion shield must be free to move in a normal direction in order to expel ice so bonding of the shield to the com- posite nose would have to be restricted to well downstream of the nose on the upper and lower surfaces. Space must also be found forward of the rotor main spar to run the required number of insulated wires to the coils as well as for the coils themselves. Also, the delicate aeroelastlc balance of the rotor must not be upset by inclusion of the EIDI system.
These concerns strongly suggest that the incorporation of EIDI into a helicopter rotor must be done as either part of a redesign of an existing rotor or design of a new system. That is, the EIDI is not an add-on system for helicopter rotor blades. Thus the development of the required technology cannot be accomplished so readily as it was for fixed wing aircraft where retrofitting of EIDI was an acceptable approach.
In the future, it is hoped that a joint NASA/Army/Industry/unlverslty consortium can be established which will develop the required technology base. It is envisioned that a series of IR1 tests will be Inltlally required followed by hover icing tests at the Ottawa Spray Rig and forward f11ght icing tests, in both artificial and natural conditions.
8o5
ICING SIMULATION
The rotorcraft icing community is severely lacking in available validated simulation techniques (both experimental and analytical) to investigate icing problems for research and development, design, or certlflcatlon/quallflcatlon purposes. A major goal of the NASA rotorcraft icing research effort is to develop and validate the required simulation techniques. The following sec- tion describes the various research efforts, the results of which will contribute to this goal.
AIRFOIL PERFORMANCE IN ICING Prior to the NASA icing research program, the vast majority of the air- foil performance-in-lcing data was acquired in the IRT by NACA researchers in the 1944 to 55 tlme period. The IRT has a maximum test section velocity of 300 mph (M = 0.4) which does not allow the high speed icing studies needed by the rotorcraft community to be conducted. It is well known that the outer span regions of helicopter rotors contribute significantly to the overall llft and drag and thus a knowledge of rotor outer span icing charac- teristics is essential. A first step to gaining thls understanding Is to understand the high speed icing characteristics of airfoil sections which are representative of outer span sections. Also, It should be noted that the airfoil geometries tested by NACA researchers in the IRT were primarily for fixed wing applications.
In order to begin to acquire the required airfoil high speed icing data base, NASA sponsored an extensive research program in the Canadian National Research Council's (NRC) 12 by 12 in. hlgh speed icing wind tunnel. At the time of this program, the NRC wind tunnel was the only operational high sped icing wlnd tunnel in North America.
This program, which was conducted by Sikorsky Aircraft and The Ohlo State University and Is discussed in reference 5, involved testing several differ- ent subscale airfoil geometries over a wlde range of aerodynamic and environ- mental conditions. Figure ll shows the main airfoil geometries tested. The NACAOOI2 airfoil represented a first generation rotor airfoil whlle the VR-7 SCI095, SCI094 R8, and SClOl2 R8 represented second generation airfoil con- tours. The SSC-A09 and circulation control sections were chosen to represent advanced airfoil designs.
A large amount of airfoil icing performance and Ice accretion shape data was acquired for Mach numbers from 0.3 to as hlgh as 0.7. In addition to airfoil force and moment levels, measurements were also made of surface pressure distributions for some of the models.
Figure 12 shows some typical data gathered during the test program.
Changes In airfoil llft and moment coefficients are shown as a function of icing time. For all airfoil configurations, increases in drag coefficient were large while the changes in llft and pltchlng moment coefficients were smaller (but certainly not insignificant).
Data such as that shown in figure 12 was used to formulate correlation equations for changes in airfoil llft and moment coefficients as functions
of aerodynamic, environmental, and airfoil geometric variables. These corre-
lation equations gave predictions which agreed with the experimental data to within about 30 percent.
These two-dlmenslonal airfoil icing correlation equations were then used along with appropriate rotor performance analysis codes to predict rotor performance in icing for the S-76, UH-IH, and UH-60A aircraft and to compare the predictions wlth available icing flight data. Predictions were made for torque rlse levels for both hover and forward flight conditions. The results are shown in figure 13. Appreciable scatter existed in the results, although the general agreement was Judged to be better for the hover condition. Gen- erally, the predicted torque rise levels for forward flight conditions were less than the observed values.
It is difficult to evaluate the shortcomings of the analysis from these limited comparisons since the quality of the flight data can be questioned.
The data was gathered during programs for which the main goal was to evaluate ice protection system performance rather than to acquire rotor performance- In-lclng data. It is felt that dedicated flight test programs must be conducted to acquire quality code validation data. Also, shedding of Ice, especially from the outer span region of the rotor, is felt to be a key aspect of rotor icing. The prediction methodology had only a very simple treatment of this complex phenomenon, and no information about shedding was contained as part of the reported experimental results. Again, future flight test programs must adequately document the shedding characteristics.
In 1985, Fluldyne Engineering Corporation (Minneapolis, MN) began opera- tlon of a 22 by 22 in. transonic icing wind tunnel (fig. 14). This is an atmospheric total pressure facility which relies on ambient conditions to provide proper alr temperature. NASA has initiated a series of tests in thls facility to acquire larger chord rotorcraft airfoil icing data. First phase of testing involved acquiring ice accretion shapes for a NACAOOI2 sec- tion for Mach numbers up to 0.6 while follow on testing will involve acqui- sition of ice shapes and resulting section drag levels. Eventually, tests will be conducted with other representative airfoil sections.
Lower speed icing studies of the NACAOOI2 airfoil (21 in. chord) have been conducted in the NASA IRT (ref 6). These studies are also aimed at acquiring airfoil ice accretion and resultant drag increase levels.
An oscillating rig mechanism for the IRT has been fabricated and will be used In future tests to assess the importance of pitch oscillation on Ice accretion and resultant aero degradation levels.
MODEL ROTOR ICING While two-dlmenslonal airfoil icing data is a required part of a compre- hensive rotor icing data base, data is also required for actual rotors operb atlng in the icing environment. This data would include Ice accretion shapes, ice shedding characteristics, and resultant rotor loads. Natural and artificial icing flight tests of full scale helicopters is the most obvious way to acquire such data, but flight testing is an expensive, man- B07 power intensive effort and It Is not realistically possible to be able to acquire data for "worst" icing conditions.
A more desirable approach to acquiring such rotor icing data would be through icing wlnd tunnel tests of properly scaled model rotors. Thls approach would allow systematic exploration within the icing envelopes of llquld water content, droplet size, and ambient temperature to determine areas of increased rotor sensitivity to icing. To date the only model rotor icing tests conducted have been done by the French (ref 7). The French claimed their results, when compared wlth natural icing flight test data, showed that meaningful Ice accretion and aero performance data could be acquired but the value of electrothermal deicer data was questioned.
Model rotor icing tests could be a valuable test technique not only for icing R&D purposes but also for icing certification purposes. However, for either application, the simulation approach must be validated by compar- ison wlth full scale flight data.
In order to develop this model rotor icing test technique, NASA, the four major helicopter companies, and Texas A&M University have formed a consortium to plan and carry out model rotor icing tests In the IRT. The first model to be tested wlll be a fully instrumented Powered Force Model (PFM) provlded by Sikorsky (fig. 15). The rotor diameter wlll be 6 ft and the airfoil section wlll be NACAOOI2. The initial tests of the PFM are ten- tatlvely scheduled for late 1987. Follow on tests would involve a properly scaled model rotor to allow direct comparisons to be made with data acquired from a complementary flight test program. It Is envisioned that thls pro- posed flight test program would likely be a Joint NASA/Army venture.
One issue which must be addressed as part of the model rotor icing effort Is what icing scaling relations to apply between full scale and model scale tests. The issue of icing scaling Is a key one within the entire icing com- munity since icing research facilities do not exist which allow testing of either full scale components or complete aircraft (model or full-scale) over the complete range of desired conditions. Current NASA efforts wlth regard to scaling are to concentrate on improving the understanding of the fundamen- tal physics of Ice accretion and to conduct comparison tests of the several scaling laws that have been published in the literature. Improvement In the formulation of Ice accretion modeling should allow improved scaling relations to be developed.
Another desirable approach to acquiring rotor aero performance-ln -icing data Is to conduct dry wlnd tunnel tests of model rotors to which have been afflxed to the leading edge representative artificial Ice shapes. Initial attempts to explore this simulation approach have involved model helicopter tests In the Texas A&M 7 by lO low speed wind tunnel (fig. lO). A two- dimensional glaze llke Ice accretion was affixed to the rotors and resulting rotor loads measured for a variety of hover and forward flight conditions.
Typical forward flight test results shown in figure 17 indicate large increases In required torque were experienced relative to the clean rotor levels. The results also indicate that the Ice shape covering the final 15 percent span of the rotor makes a large contribution to overall performance degradation.
Supporting two-dlmenslonal airfoil tests were conducted in the Texas A&M 7 by lO wind tunnel to measure local section performance degradation due to the ice shape. Overall results of both the two-dlmenslonal airfoil and model rotor tests are reviewed in reference 8.
Follow on testing of this model helicopter is currently underway to investigate the importance of using more detailed artificial ice shapes.
Simulated asymmetric ice shedding cases are also being studied.
The artificial ice shape testing approach offers another potentially attractive simulation testing technique if it can be properly validated.
The ice shapes could be determined by analytical predictions, available experimental data and/or personal Judgement. Data acquired could be useful for validating analytical prediction methodologies and for providing data to support quallficatlon/certlflcatlon programs.
FULL SCALE HELICOPIER ICING As previously indicated, icing flight testing is a very expensive and time consumlng test technique, and validated s_mulatlon approaches are pre- ferable. However, some icing flight testing is required to develop the necessary validation data base. These test programs must be dedicated to acquiring such data rather than the data being acquired as a by product of a program aimed at some other goal, such as, development of an ice protection system.
NASA and the Army have completed the first such flight program called the Helicopter Icing Flight Test (HIFT) program (refs. 9 and lO). The over- all HIFT program has been a multi year effort which has involved several other organizations Bell Helicopter Textron, Fluldyne Engineering Corporation, Hovey and Associates, and The Ohio State University.
The major elements of the HIFI program and how they are related are shown in figure 18. Phase I flight tests were conducted in the Canadian NRC'S Ottawa Spray Rig while Phase II flight tests were conducted behind the Army's HISS tanker. For each Phase of the flight testing, ice was allowed to accrete on the unprotected rotors of the test aircraft, a UH-IH, and result- ant aero performance degradation was measured. The test aircraft was then landed with care so as to minimize the ice shed from the rotors during shut down. Rotor ice accretions were then measured along the span primarily using two techniques - ice shape tracings and silicone molds.
Ice shape tracings recorded along the span of the rotor for flight E of the Phase I hover tests are shown in figure 19 while figure 20 shows one of the silicone molds acquired for the same flight. The increase In horsepower required to sustain hover flight for the icing of flight E was 96-HP which was about a 17 percent increase over the clean rotor value. Summaries of the Ice shape documentation efforts for the two phases of flight testing are given in references II and 12.
The slllcone molds acquired from Flight E were then used to make detailed epoxy castings of the ice accretions such as the one shown in figure 21.
Full scale UHIH rotor sections, with these castings affixed to the leading C, edge were tested in the Fluidyne dry transonic wind tunnel to measure rotor sections drag increases due to the artificial ice shapes. The results of these tests are shown in figure 22. These results were then used along with Bell rotor performance codes to predict the rotor degradation. The compar- isons for Flight E shown in figure 23 indicated good agreement existed between measured and predicted performance degradation.
For the forward flight tests (phase II), one flight test point has been chosen for follow-on testing and analysis. The wind tunnel tests have been completed and the rotor performance degradation predictions will soon be made.
The HIFT program has demonstrated that it is possible to acquire rotor ice accretlon/aero-performance data for hover and forward flight icing.
Future test programs should incorporate In-fllght ice shape documentation in addition to the ground documentation techniques employed in the HIFT program.
one such flight system might be the hub mounted video camera system developed by the Army as part of the rotor pneumatic boot program already discussed.
Such a system could document when and where ice shedding occurred during the test flights.
ROTOR ICING ANALYSIS The major efforts in rotor icing analysis are in the areas of rotor aero performance degradation and electrothermal de-lclng performance.
Initial aero performance predictions were accomplished through Joint NASA/Boelng-Vertol/Texas A&M studies to predict Boeing Chinook front rotor degradation due to rime icing (refs. 13 and 14). The intent of this initial study was to determine whether it was feasible to predict rotor performance in icing using a purely theoretical approach. In particular, a droplet tra- Jectory code, an airfoil performance degradation correlation expression, and a rotor performance code were coupled to form an analysis methodology.
Initial studies were done for hover icing (ref. 13) and then extended to forward flight conditions (ref. 14). The droplet trajectory code used was that developed by Bragg (ref. 15).
Figure 24 shows the predictions for horsepower required to maintain a constant value of rotor thrust coefficient for two different assumed radial extents of icing (85 and lO0 percent). The figure show the important contri- bution of the rotor outer span to overall rotor degradation. These results clearly show that a comprehensive rotor icing performance methodology must be able to predict radial extent of icing. Also these results suggest the importance of being able to handle rotor ice shedding, especially in the outer span region.
The hover icing predictions were encouraging enough to extend the predic- tions to forward flight conditions. This conclusion was primarily on the basis that the performance degradation levels predicted were reasonable, especially when compared to reported values for other helicopters. Unfortu- nately, no data was available at the time for Chinook performance in icing.
For the hover icing analys_s, It was possible to take advantage of the
rotor axlal symmetry so that droplet trajectory and alrfo11 performance anzlyses had to be performed only as a function of radial position and not also as a function of azimuthal position. However, for the forward fll_ht analysis It was necessary to account for the azimuthal dependence. _ .......
Initially, trajectory and airfoil performance calculations were perfoi_med for 13 radial locations for each 15 o increment of rotor disk (24 total azimuthal locations). Thts matrix of information was then input Into the forward fltght performance code and rotor degradation predicted. A simpli- fied approach was also investigated which tnvolved calculating average sec- tton Mach number and angle-of-attack for each radlal location on the rotor.
These average conditions were then used as inputs Into the trajectory and alrfoll performance analyses. The resultant array of atrfoll degradation levels were then input Into the rotor performance code to make the flnal predlctlon. While the number of calculations required for this simplified approach was far less than for the detailed analyses, the results differed by no more than 5 percent In predictions of horsepower required to sustain level flight.
Figure 25 shows the predictions of horsepower required (referenced to the no Ice value) using the detailed analytical approach. Four different assumed radial extents of icing are shown, and again the importance of the outer span icing ts evident. Sikorsky employed thts same basic analytical approach In the studies already mentioned. Improvements on the Initial NASA/ Boetng-Vertol/Texas A&M methodology were In the areas of predictions of spanwtse extent of tclng, prediction of Ice shedding, and use of general alrfoll correlations rather than Just a rime icing correlation.
The major conclusion to be reached from the rotor icing analysis efforts completed Is that It appears to be possible to develop a comprehensive rotor icing analysis methodology which wlll predict wlth reasonable accuracy the rotor degradation levels to be expected from Ice accretions. Current NASA efforts are aimed at developing such a comprehensive methodology, the maln elements of which are shown In figure 26. In the near term, correlations w111 be used to predlct alrfo_l degradation due to _clng while in the longer term It Is envlsloned that these correlations wlll be replaced by codes whlch predict the actual Ice accretion distribution on the rotor and the resultant section changes In lift, drag, and moment coefflclents. These codes are currently being developed and validated by NASA as part of the aircraft Iclng research program. Also, ongoing fundamental research (both analytical and experimental) In the area of Ice structural properties should provlde some guidance In the development of the Ice shedding module.
While this rotor icing analysis methodology wlll certainly require sig- nificant computational resources, It Is felt that It w111 be a fundamentally correct methodology which will not be based on extensive use of correlations which can always be questioned when "new" or different rotor configurations are Investigated.
Again, It must be emphasized that additional high quallty icing flight test data must be acquired to validate this rotor icing analysis methodology.
Currently, the data being used Is that provided through the HIFT program as well as unprotected Brltlsh RAF Chinook data provided by the British Royal
Aircraft Establishment (RAE) through the Joint NASA/RAE studies on helicopter
Icing.
The other analysis effort has been the development and validation of a series of transient heat conduction codes which model the electrothermal delcer. Figure 27 gives the important characteristics of the slx different codes developed to date and how they differ from one another. As the table suggests, the codes vary greatly in complexity and therefore, in the size of computer required. Codes 1 and 2 which model the electrothermal deicer on a one-dlmenslonal, time varying basis can be run on a personal computer.
Code 6 which analyzes the complete two-dimensional geometry, Including the varlable thickness ice layer and the many layers of the airfoil geometry requlres a supercomputer (i.e. a Cray 1S or XMP) to achieve reasonable run times. A unique feature of all codes is that they treat the melttng of the ice through a phase change routine.
Currently, the code predictions are being compared with electrothermal deicer data from icing wind tunnel and helicopter natural icing flight-tests.
These comparisons are aimed at providing a better understanding of how accur- ately the various codes can predict electrothermal deicer performance for the various levels of modeling complexlty.
References 16 to 20 give detalled descriptions of the Individual electro- thermal deicer codes.
CONCLUDING REMARKS This paper has reviewed the accomplishments to date of NASA research efforts related to helicopter icing. Some suggestion has also been given as to future planned efforts. The two major areas of concentration will con- tinue to be developing advanced rotor Ice protection concepts and developing/ validating rotor icing analytlcal and experimental simulation methodologies.
The development of advanced ice protection concepts will provide the helicop- ter industry with alternatives to the electrothermal system which is viewed by many to be two expensive and too heavy and to require too much power.
Validated analytical and experimental simulation methodologies are tools needed by the industry to increase their understanding of the rotor icing problem and to reduce time and cost associated with icing certlflcatlon/ qualification. In order to accomplish a timely transfer of this icing tech- nology, most of the research efforts are performed Jointly with industry through either formal contracts or co-operatlve programs.
REFERENCES I. Werner, J.B.: Ice Protection Investigation for Advanced Rotary-Wing Aircraft. LR-25327-I0. Lockheed-Callfornla Co., 1973. (USAAMRDL-TR- 73-38, AD-771182) 2. Blaha, B.J.; and Evanlch, P.L.: Pneumatic Boot for Helicopter Rotor Deicing. NASA CP-2170, 1980.
3. Haworth, L.A.; and Graham, M.S.: Fltght Tests of the Helicopter Pneuma- tic Deicing System. American Helicopter Society. National Specialists Meeting, Helicopter Testing Technology, October 29 - November l, 1984.
4. Zumwalt, G.W.; Schrag, R.L.; Bernhart, W.D.; and Frledburg, R.A.: Analysis and Tests for Design of an Electro-Impulse Deicing System.
NASA CR-174919, May 1985.
5. Flemmlng, R.3.; and Lednicer, D.A.: Hlgh Speed Ice Accretion on Rotorcraft Airfoils. NASA CR-3910, August 1983.
6. Olsen, W.; Shaw, R.; and Newton, J.: Ice Shapes and the Resulting Drag Increase for a NACAOOI2 Airfoil. NASA TM-83556, 1984.
7. Guffond, d.P.: Icing and Deicing Test on a I/4 Scale Rotor In the ONERA SIMA Wlnd Tunnel. AIAA Paper 86-XXX, January 1986.
8. Korkan, K.D.; Cross, E.J.; and Cornell, C.C.: Experimental Study of Performance Degradation of Performance Degradation of a Model Helicopter Maln Rotor wlth Simulated Ice Shapes. AIAA Paper B4-0184, January 1984.
Evaluation 9. Abbott, W.Y.; Belte, D.; Williams, R.A.; and Stellar, F.W.: USAAEFA of UH-1H Hover Performance Degradation Caused by Rotor Icing.
Report 82-12, August 1983.
lO. Abbott, W.Y.; Linehan, J.L.; Lockwood, R.A.; and Todd, L.L.: Evaluation of UH-IH Level Flight Performance Degradation Caused by Rotor Icing.
USAAEFA Report 83-23.
II. Lee, J.O.; Harding, R.; and Palko, R.: Documentation of Ice Shapes on the Main Rotor of a UH-1H Helicopter In Hover. NASA CR-168332.
12. Hanson, M.K.; and Lee, J.D.: Documentation of Ice Shapes on the Main Rotor of a UH-IH Helicopter In Level Flight. NASA CR-175088, March 1986.
13. Korkan, K.D.; Dadone, L.; and Shaw, R.3.: Helicopter Rotor Performance AIAA Journal of Aircraft, Degradation on a National Icing Encounter.
Vol. 21, 1984, p.84.
14. Korkan, K.D.; Dadone, L.; and Shaw, R.J.: Performance Degradation of Helicopter Rotor Systems In Forward Flight Due to Rlme Ice Accretion.
AIAA Paper 83-0029, 19B3.
15. Bragg, M.B.: Rlme Ice Accretion and Its Effect on Airfoil Performance.
NASA CR-165599, 1982.
16. DeWitt, K.J.; and Ballga, G.: Numerical Simulation of One-Dlmenslonal Heat Transfer In Composite Bodies wlth Phase Change. NASA CR-165607, 1982.
l?. Marano, J.J.: Numerical Simulation of an Electrothermal Deicer Pad.
NASA CR-168097, 1983.
18. Chou, O.F.K.: Numerlcal Simulation of Two-Dimensional Heat Transfer tn Composite Bodies with Application to Deicing of Aircraft Components.
NASA CR-168283, 1983.
19. Naslulanlec, K.K., et al.: Full Two-Dlmenslonal Translent Solutions of Electrothermal Aircraft Blade Deicing. AIAA Paper 85-0413, January 1985.
20. Leffel, K.L.: A Numerlcal and Experimental Investigation of Electro- thermal Aircraft Oetctng. NASA CR-175024, 1986.
_Ooo t I FIGURE 2. - PNEIJRATIC BOOT ON ROTOR SECTION RODEL INSTALLED IN 6x9 FT NASA LEWIS ICING RESEARCH TUNNEL.
, • - = 'T t; |
t.! v
f
_ _-'/
,,J II
_DD
.,J Z [t "_ , P
'i
t
t'
N U.
_1 I I I I I I I I a:) "111:11:)I:1:130:) 9V11_ NOIJI.:)_S PNEUMATIC DEICER J" EJECTOR FLOWCONTROL-', VALVE REGULATOI RELIEF VALVE _ CHECKVALVE TURBINE BLEED 1 _L LOW-VACUUM WARNING SWITCH FIGURE 5. - MAIN ROTOR APPLICATION OF PNEUMATIC DEICER.
FIGURE 6.- CANADIAN NATIONAL RESEARCH COUNCIL'S OTTAWA SPRAY RIG.
FIGURE 7,- ARRY'S HELICOPTER ICING SPRAYSYSTER (HISS) TANKER.
] [ --/ \\\ /"_/ W,.--TO SWITCHING ALURIllUR DOUBLER .-_" _ III CON_LED COIL"/ FIEURE 8,- EIDI IMPULSE COILS IN A LEADING EDGE.
FIGURE 9.- AHI CORBA ROTOR SECTION IN NASA IRT WITH LEADING EDGE ICE ACCRETION.
82O O.I_IG_;_JAL PAGE ;_ DF. POOR QUALIT_ FIGURE I0.- AHI COBRA ROTOR SECTION AFTER ACTIVATION OF EIDI SYSTEM.
000 (1 i
'ill
F ---___
,I I I I I I I AIRFOIL NACA 0012 SCI095 (LWC = 1.03 G/M) VR-7 / SCI012R8
/
Moo = 58, a = 6 / /, LWC = O. 94 G/M 3 / ' / /, / • /
/,/ /
/// / #@ z // / os
/
I I I I I _ _ ',._- 02 | -.o. i I i I -"'1 0 20 40 60 ICING TIME, SEC FIGURE 12.-EFFECT OF ICING TIME AT AN ANGLEOF ATTACK OF B DEGREESAT HIGH MACH NUMBERS.
3O LINE OF PERF AGREERENT_ 2O 0 0 / D_ _ °z_ n / 0 s-76'81 / D s-76 '82 -- / _ UH-1H V UH-GOA
2 _ _ _
_ OOD [] V" o I I I 0 I ,.=, (A) HOVER.
2O
-
_vvnv I v v v I I 10 20 30 OBSERVEDTOROUERISE, PERCENT (B) FORWARD FLIGHT.
FIGURE 13.- CORPARISONOF PREDICTED AND REASURED TOROUE RISES FOR HOVER AND FORWARD FLIGHT ICING CONDITIONS.
_0_ _POOR _Q[_L_IT_3 FIGURE 14,- FLUIDYNE HIGH SPEED ICING WIND TUNNEL.
FIGURE 15,- SIKORSKY POWERED FORCE MODEL TO BE TESTED IN NASA IRT.
FIGURE 16.- MODELHELICOPTER INSTALLED IN TEXAS A & M 7x10 LOW SPEED WIND TUNNEL.
.... _ f r_ r'T_ " ©_ POOR QUAL_ 7 m 6 -- % x q 1 I I I I 2 3 q 5 6 CT x 103 FIGURE 17.- VARIATION OF TORgUECOEFFICIENT VERSUS THRUST COEFFICIENT FOR VARIOUS SPAI_ISE ADDITIONS OF GENERIC ICE: FORWARD-FL ]GHT CONDITIOf(.
O3
!
=
!"t
i Z = .O_ POOi_ QUALIT'_ R/R .2 .3 .q .5 .6 .7 .8 .9 1.0 I I I I I I I I I 0.05 x/c EXTENT OF ICE ACCRETION-", \ FIGURE 19.- SPANWlSE VARIATION IN ROTOR ICE SHAPES FOR FLIGHT E OF PHASE I TESTING.
FIGURE 20.- TYPICAL ROLDS OBTAINED OF ICE FORRATION FOR FLIGHT E.
/ .....
C-83-7031 FIGURE 21.- EPOXYCASTING AFFIXED TO UH-1H ROTORSECTION.
O ClIAN "ICED" R/R = 0.44 R/R = 0.60 R/R = 0.77 .1 m A
aa
A 0 _ _ A_
o oo _ _oo _
.ol 0 0
I [____1__ I I
.001 ___ I 0 5 10 0 5 0 S ANGLE-OF-ATTACK, DEG FIGURE 22,- DRY WIND TUNNEL RESULT FOR "ICED" AIRFOIL PERFORMANCE.
100 -- 95 -- ,?, 90 -- 85 -- 8O FLIGHT TEST HOVER FORWARD EXI RAPOLATION FL IGflT ANALYSIS PREDICTIONS FIGURE 23.- UH-1H ROTOR PERFORMANCE DEGRADAIION DUE TO ICING FOR FLIGHT E - EXPERIMLNTAL AND PREDICTED.
i I I I I I I I I 40 80 120 160 200 240 280 300 lIME, SEC FIGURE 24.- HORSEPOWER REQUIRED TO MAINIAIN A C T OF 0.004 AS A FUNCTION OF ICING TIME - CH47D HELICOPTER ROTOR BLADE, HOVER CONDITION.
O _" i Z _t _ e o a_ i..- i_ , o =___ g ,_ _ - _ I z N u.
O N ('d3_)dH/(331)dH LL ,.z LJ °