Document
- IV tt~tt -,~, -u , c;-- -
NASA Technical Memorandum 87265
AIAA-86-9758
NASA-TM-87265
, I q ~ ~ 00/ ?; 1//
\_--------~
The Measurement of Aircraft Performance
and Stability and Control After Flight
Through Natural Icing Conditions
Richard J. Ranaudo, Kevin L. Mikkelsen, Robert C. McKnight,
Robert F. Ide, and Andrew L. Reehorst
Lewis Research Center
Cleveland, Ohio
and
Jerry L. Jordan, William C. Schinstock,
and Stewart J. Platz
Kohlman Systems Research, Inc.
Lawrence, Kansas
JUL 1 1986
l,hNGLEY RESEARCH CENTER LIBRARY, NASA HAI.1PTON, VlRGINJf\
Prepared for the
3rd Flight Testing Conference
cosponsored by the AIAA, AHS, CASI, DGLR,
IES, ISA, ITEA, SETP, and SFTE
Las Vegas, Nevada, April 2-4, 1986
111111111111111111111111111111111111111111111 .
NF01511
NI\S/\
AIAA-86-9758
The Measurement of Aircraft Performance
and Stability and Control After Flight
Through Natural ICing Conditions
Richard J. Ranaudo, Kevin L. Mikkelsen,
Robert C. McKnight, Robert F. Ide, and
Andrew L. Reehorst, Lewis Research
Center, Cleveland, OH; and Jerry L.
Jordan, William C. Schinstock, and Stewart
J. Platz, Kohlman Systems Research, Inc.,
Lawrence, KA
·
,
AIAAI AHS/CASI/DGLR/IES/ISA/ITEA/SETP I
..
SFTE 3rd Flight Testing Conference
April 2-4, 1986/Las Vegas, Nevada
For permission to copy or republish, contact the American Institute of Aeronautics and Astronautics
1633 Broadway, New York, NY 10019 NSGa "'~@51S&=t±-
THE MEASUREMENT OF AIRCRAFT PERFORMANCE AND STABILITY AND CONTROL AFTER FLIGHT THROUGH NATURAL ICING CONDITIONS Richard J. Ranaudo, Kevin L. Mikkelsen, Robert C. McKnight, Robert F. Ide, . and Andrew L. Reehorst National Aeronautics and Space Administration Lewis Research Center Cleveland, Ohio 44135 and Jerry L. Jordan, William C. Schinstock, and Stewart J. Platz Kohlman Systems Research, Inc.
Lawrence, Kansas 66044 Abstract change in aircraft lift coefficient with elevator deflection (elevator The effects of airframe icing on the performance effectiveness) per deg and stability and control of a twin-engine commuter-class aircraft were measured by the NASA aircraft pitching moment, untrimmed Lewis Research Center. This work consisted of clear air tests with artificial ice shapes attached aircraft static pitching moment to the horizontal tail, and natural icing flight parameter, per deg tests in measured icing clouds. The clear air tests employed static longitudinal flight test + C .) MMLE pitch damping state coefficient, m methods to determine degradation in stability a per rad margins for four simulated ice shapes. The natu- ral icing flight tests employed dynamic flight change in aircraft pitching moment with maneuvers and a compatible data acquisition system, elevator deflection (elevator power) which was provided under contract to NASA by per deg Kohlman Systems Research Incorporated. This system used a performance modeling method and thrust coefficient modified maximum likelihood estimation (MMLE) technique to determine aircraft performance aircraft normal force coefficient degradation and stability and control. Flight obtained by g x W/q x S test results with artificial ice shapes showed that longitudinal, stick-fixed, static margins elevator control force normalized by are reduced on the order of 5 percent with flaps dynamic pressure up. Natural icing tests with the KSR system cor- roborated these results and showed degradation in FSSP forward scattering spectrometer probe the elevator control derivatives on the order of 8 to 16 percent depending on wing flap configura- g normal acceleration due to gravity tion. Performance analyses showed the individual contributions of major airframe components to the KCAS calibrated airspeed, knots overall degradation in lift and drag.
LWC liquid water content in cloud, gm/m Performance modeling methods and MMLE tech- niques are viable flight test methods for deter- M Mach number mining the effects of natural ice on aircraft performance and stability and control. These MAC mean aerodynamic chord of the wing, ft techniques have an advantage over static methods because they provide for a rapid acquisition of MMLE modified maximum likelyhood estimation flight test data in an environment where test technique time is constrained by the rate that ice shapes sublimate, melt, or erode. Measurements of sta- MVD median volume diameter of droplets, ~m bility and control with MMLE are limited to those portions of the flight envelope where aircraft S wing area, ft2 response remains essentially buffet-free.
SHP shaft horsepower Symbols and Definitions W aircraft weight, lbs CD aircraft drag coefficient C trimmed aircraft lift coefficient a aircraft angle of attack, deg L aircraft yaw angle, deg (C + CLa)MMLE derived state coefficient, per rad Lq ambient static pressure divided by C aircraft lift curve slope, per deg static pressure at standard day, sea L a level conditions This paper Is declared a work of the U.S. Government and Is not subject to copyright protection In the United States.
0a aileron deflection, deg Aircraft modifications for the static longi- tudinal tests with artificial ice shapes were 0e elevator deflection, deg accomplished in house. Data from these flights were also reduced and analyzed in-house. Modifi- of flap deflection, deg cations to the aircraft for the transient response flight tests and the methodology for data reduc- or rudder deflection, deg tion and analysis were provided under contract to NASA Lewis by Kohlman Systems ResSarch, Inc. (KSR), ~ denotes change in a parameter Lawrence, Kansas. The KSR system is capable of rapidly acquiring aircraft performance and sta- Introduction bility and control data with a minimal number of flight test maneuvers. This system, described The purpose of this paper is to present later in this report, uses performance modeling quantitative data showing the effects of icing on methods and modified maximum likelihood estimation the performance and stability and control of a (MMLE) techniques to perform an analysis of twin-engine, commuter-class aircraft. These data, dynamic aircraft maneuvers and calculate aero- which were acquired solely through flight testing, dynamic coefficients and stability derivatives.
provide a unique association between measured This system, however, had never been used on an icing cloud properties and the effect these prop- aircraft whose aerodynamic characteristics were erties have on aerodynamic coefficients and sta- altered by icing. Thus, one of the peripheral bility derivatives. This paper also presents the obje~tives of this program was to determine the methods by which flight data were acquired and utility of performance modeling and MMLE tech- discusses the application of transient response niques for natural icing flight tests.
flight test techniques for aircraft whose basic aerodynamics are altered by ice formations.
Only a portion of the data acquired with the KSR system was analyzed for this paper. Priority All aircraft are susceptible to icing when was given to an analysis of aircraft performance, flying through clouds that contain super-cooled longitudinal stability and control. Lateral- water droplets. The amount and type of ice form- directional data will be analyzed and presented ing on forward-facing aircraft surfaces is a func- in a follow-on report.
tion of several variables that include aircraft speed, cloud liquid water content, temperature, The Research Aircraft water droplet size distribution, airfoil or body geometry, angle ~f attack, and duration of the The icing research aircraft is a modified icing encounter. The effects of icing on air- DeHavilland DH-6 Twin Otter. This aircraft, as craft performance were measured by the NASA Lewis shown in Fig. I, is powered by two Pratt Whitney over a range of natural icing conditions as PT6A-20A gas turbine engines that deliver 550SHP reported in Refs. 2 and 3. These measurements each at standard day, sea level conditions.
were made by employing stabilized level flight Aircraft weights, inertias, and dimensions are performance methods due to the limitations imposed provided in Table I.
by the type of flight instrumentation and data acquisition systems installed in the aircraft at Research Instrumentation Systems that time. The data from these tests are cur- rently being applied to the d~velopment of com- Icing Instrumentation and the Wing Stereo Camera puter codes that pre~ict aircraft performance System losses due to icing.
The icing research aircraft measures cloud Experience gained during these earlier properties with several instruments as shown in flights indicated a need for more expedient per- Fig. 2, and the operation of these instruments is formance flight testing techniques to reduce the described in Ref. 6. To eliminate redundancy, effects of ice shape deterioration on the data.
data only from three icing instruments were used Accreted ice shapes, for example, were found to to document icing cloud properties: liquid water deteriorate through sublimation, melting, or ero- content (LWC) values were derived from the sion; and the test time available before these Rosemount Ice Detector; median volume droplet effects became significant depended largely on diameters (MVD) from the Foreward Scattering temperature, sunlight conditions, and flight speed.
Spectrometer Probe (FSSP) produced by Particle In a parallel development, NASA Lewis expanded its Measuring Systems; and free air temperature from icing research program to include an investigation the Rosemount total air temgerature probe. Stereo of the effects of icing on aircraft stability and analyses of wing ice shapes were provided by control. This new initiative, along with the the Arnold Engineering and Development Center requirement to improve aircraft performance data (AEDC) from in-flight photographs (stereo pairs) acquisition techniques, resulted in the formula- taken with the stereo camera system (Fig. 3).
tion of a flight program consisting of three major elements: static longitudinal stability and con- Instrumentation for Performance and Stability and trol fl ight tests in clear air with·artificial Control Measurements ice shapes to gain experience with aircraft characteristics and help provide definition for The iCing research aircraft is equipped with natural icing tests; performance flight tests in control position transducers, a dynamic force natural icing conditions employing level acceler- wheel, and rudder pedal load cells as shown in ation and deceleration techniques; and, stability Fig. 4. Airspeed; altitude, angle of attack (as and control flight tests in natural icing condi- referenced to the fuselage floor line), and angle tions employing transient response flight test of sideslip were measured by a heated Rosemount techniques.
858 probe as shown in Fig.5. This probe was flight calibrated for static source error by the Once the shape was determined, aluminum tem- trailing static cone method. Angle of attack plates were fabricated and artificial "ice" was cut from styrofoam blocks with a hot wire. The was calibrated in flight with a floor-mounted inclinometer, and angle of sideslip was likewise artificial ice shapes were then glued to a thin, flight calibrated against an inertially derived full-span, aluminum leading edge cap that con- sideslip from the KSR gyro package. Control posi- formed to the leading edge radius of the horizon- tions and forces were calibrated by comparing tal tail plane. The aluminum cap was secured by known inputs with recorded sensor outputs. Con- chordwise straps and clamped to the stabilizer trol position extremes and zero reference forces hinge line. A typical installation for the were recorded before and after each test flight. moderate glaze ice shape is shown in Fig. 6(d).
During static longitudinal flight testing with Prior to flight testing, a basic operating artificial ice shapes all flight parameters were recorded on a digital tape recorder and later weight and center of gravity (c.g.) was obtained reduced to engineering units for analysis. by weighing the aircraft with all flight test instrumentation installed less fuel and crew.
KSR Data Acquisition System Individual crew members were then weighed, and their weights and seating locations were used to Aircraft performance and stability and con- calculate a zero fuel weight and balance. It was trol for icing flights were measured by a modular found that by varying the seating-locations of the cabin crew members, three evenly spaced c.g.'s data acquisition system (DAS) provided by KSR.
could be obtained between 26 and 34 percent of The KSR DAS is a 12-bit system that records approximately 8.6 samples per second with all the MAC. This rather expeditious procedure of channels sampled within a 1 msec time interval varying c.g. in flight was then used to provide and then recorded on tape. This effectively the necessary range of c.g.'s necessary for static eliminates time skews due to sequential sampling. longitudinal flight test maneuvers. Fuel weights The DAS includes the computer, computer controls recorded at the beginning and end of each maneuver and display, tape recorder, and Signal condition- were applied to the zero fuel weight calculation ing. The parameters recorded by this system for to obtain the average test weight and c.g. for the performance and stability and control measurements maneuver.
are listed in Table II.
Classical longitudinal flight test maneuvers Flight Testing consisting of sawtooth climbs and descents and wind-up turns were employed to determine stick- Static longitudinal flight test methods were fixed and stick-free neutral and maneuver points.
employed to examine the effect of artificial ice All sawtooth maneuvers were flown at a nominal cruise power of 275 SHP/engine. Wind-up turns ~hapes on stability margins. Transient response were accomplished at 120 KCAS and 103 KCAS. The methods were employed to examine aircraft perform- test methods used to acquire these flight data ance and stability and control degradation after flight through natural icing conditions. Clear are found in Ref. 9. With artificial ice shapes attached to the tail, special attention was given testing was required for both the static longi- tudinal and transient response methods to define to the takeoff condition. Since no crew escape baseline coefficients and derivatives for the systems were available and the artificial ice shapes were not jettisonable, testing was clean aircraft.
approached very conservatively. All flights Clear Air Flight Testing were thus restricted to the flaps-up configura- tion. During takeoff with each ice shape, nose- wheel unstick speeds were noted relative to the Static longitudinal flight testing with baseline configuration; and positive stable artificial ice shapes attached to the horizontal response was verified through rotation and tail. - An abbreviated flight test program with liftoff.
artlficial ice shapes attached to the horizontal tail was conducted prior to the installation of KSR performance and stability and control the KSR system to help identify how aircraft han- dling qualities and stability derivatives might tests. - A complete performance and stability and be expected to change with natural ice accretions. control (MMLE) baseline (noniced) was derived for This test program consisted of five research the Twin Otter aircraft prior to the conduct of flights. One flight established a "no-ice" base- icing research flights. This baseline was line data set while the remaining flights were acquired by conducting clear air tests with the flown with four characteristic ice types that aircraft configured as it would be for iCing simulated: surface roughness (initial stage of flight, i.e., with all externally mounted probes an icing encounter); rime ice (Fig. 6(a)); light and sensors in place.
glaze ice (Fig. 6(b)); and moderate glaze ice (Fig. 6(c)). The surface roughness condition Transient response flight testing did not (not shown here) was simulated by wrapping the require maneuver repetition at different center leading edge of the horizontal tail with 50 grit of gravity locations, but a very accurate air- sandpaper over an assumed impingement area. The craft weight and center of gravity had to be known simulated ice shapes in Figs. 6(a) to (c) were for each test maneuver. Therefore, weight and balance was obtained for each baseline flight by based upon photographs taken of the Twin Otter ."
tail on previous natural icing research flights. weighing the aircraft with all research equipment Environmental data from these flights were used on board before and after fueling. Actual crew in combinafion with the ice shape calculation member weights and seating locations along with measured fuel burn off were then applied to procedures and the respective two-dimensional photograph to establish a cross sectional shape determine aircraft weight and balance for each for testing.
test point flown. The procedure of weighing the aircraft before and after fueling was not used that day; the required iced configurations obtain- for icing research flights since fuel weight cal- able through selective deicing; and the desired culations were adequately validated on the base- flap settings defined for a given set of data points.
line flights.
Baseline performance flight test maneuvers Winter weather conditions in the Great Lakes region provide NASA Lewis with an excellent geo- consisted of level, full and partial power accel- graphical location for natural iCing research.
erations and decelerations, with a few stabilized "Lake effect" clouds formed by rising moist air points flown at selected airspeeds. The perform- ance baseline was obtained in the cruise and non- over northeastern Ohio and western Pennsylvania cruise (wing flaps extended) configurations. provide an environmental situation where a high probability of icing exists even when it is not Throughout all performance maneuvers a constant forecast. Lake effect clouds are generally weight to pressure (W/ ratio was maintained; o) that is, as fuel was burned off, each test maneu- stratoform-type clouds with tops below 10 000' MSL.
ver would be flown at successively higher altitude This condition is ideal for the Twin Otter since increments. Thus, a constant W/o would be it is not pressurized and use of crew oxygen is maintained from the arbitrary starting altitude. seldom required. It is also advantageous since the higher engine power available below 10 000' Baseline MMLE flight test maneuvers consisted allows greater power margins for intentional air- of longitudinal doublets, lateral directional frame ice buildup.
doublets, and asymmetric power sideslips. Longi- Flight planning for icing research flights tudinal doublets were performed in each of four wing flap configurations, i.e., 0°, 10°, 20°, and began with a systematic check on area forecasts, 37.5°. A range of four speeds were selected for hourly weather observations, synopses, upper air each configuration to provide a reasonable spread soundings, terminal forecasts, and most import- of angles of attack without encountering buffet antly, current pilot reports. In many cases, the during the 1.5 g positive acceleration portion of FAA air traffic control facility in charge of the doublet. Each maneuver began with power set airspace where icing was reported or forecast for level flight and the aircraft in trim. Longi- would be contacted by telepho~e and asked to fur- tudinal doublets were initially attempted at ther query aircraft in their sector for tempera- speeds as low as 110 percent of stall speed for a ture, sky condition, and icing. In many cases, given weight and configuratio~. It was found this procedure resulted in a successful icing through analysis that moderate stall buffet asso- encounter whereas reliance on only weather and/or ciated with the positive (1.3 to 1.5 g) accelera- available pilot reports would have resulted in a tion portion of the doublet would not allow MMLE mission cancellation.
to converge. Thus, the low-speed, longitudinal doublet maneuvers were restricted to speeds that Enroute to the known icing area, an attempt would be made to fly in the clear or at an alti- would provide only very light buffet at approxi- mately 1.5 g acceleration. This restriction gen- tude where no ice accretions could inadvertantly erally required some experimentation on the part form on the aircraft. The purpose for this was of the pilot to determine the slowest speed to eliminate extraneous ice accretions on the attainable as a function of test weight on base- airframe, which would not be considered represen- tative of the measured icing encounter.
line flights. For icing flights, this speed was found to be around 10 to 15 knots higher based Testing with naturally accreted ice shapes upon test weight and the type and amount of ice on the wing. required some special considerations. Upon exit- ing icing conditions, the aircraft would be flown All baseline longitudinal maneuvers were so as to climb above or descend below the icing also flown at three discrete power settings, i.e., cloud. If testing was to be perfomed without maximum continuous power, power for level flight, cloud cover obscuring the sun, an attempt would and an approximate zero thrust condition. These be made to stay out of direct sunlight by placing data enabled the isolation of power effects over the sun at the "six o'clock" position while flight the range of speeds tested when the aircraft was test maneuvers were being performed. Also, the iced up. Baseline MMLE maneuvers were also test matrix would be flown such that most slow repeated at two altitudes (approximately 7000' speed data points would be accomplished first to and 13 000' MSL). further reduce the amount of ice erosion. Expedi- tious accomplishment of flight test maneuvers was Icing Research Flight Testing another very critical factor in completing the test matrix before ice shape deterioration became sig- The basic approach used in the icing flight nificant. This required a very well-coordinated, research experiment was to fly the aircraft organized, and concentrated effort on the part of through icing clouds with all ice protection the pilot and flight test engineer. While flight systems (except engine inlet and propeller test maneuvers were being flown, other crew mem- heaters) turned off. Icing cloud properties were bers recorded wing stereo photography, photo docu- continuously measured and recorded while the air- mentation of ice shapes on nonlifting surfaces, pilot comments, and flight clearances. Generally craft was maintained at a nominal cruise airspeed.
speaking, a complete aircraft performance test Once a sufficient amount of ice had formed, the aircraft would be flown clear of the clouds and including level accelerations and decelerations stereo photographs taken of the ice shapes, which to break out drag and lift decrements due to ice had accreted on the right outboard wing panel. on the wings, horizontal tail, vertical tail, Dynamic flight maneuvers appropriate to perform- wing and landing gear struts, and the combined ance or stability and control testing were then remainder of nonprotected components (radome, flown in a predetermined sequence. The sequence antennae, flap hinge brackets, etc.) was accomplished within 20 to 25 min after the icing was established by: the mission objectives for encounter. For this period of time, ice accre- it to engineering units, and storing it on a disk tions would usually retain their original shapes and tape. The data then entered a preprocessor fairly well.
program in which corrections were made regarding position error of the airspeed system and location Stability and control testing was more of the accelerometers away from the center of involved. Here both longitudinal and lateral gravity. Also computed were the weight, center directional maneuvers had to be accomplished in of gravity, moments and products of inertia, several prep1anned "iced" configurations. For engine parameters (thrust, fuel flow, RPM, etc.)
reasons of flight safety, longitudinal maneuvers and other derived parameters that were needed for were performed only in the zero and ten degreee two or more of the succeeding analysis programs.
flap condition. Longitudinal and lateral direc- Both the corrected and derived parameters were tional doublets would also be repeated for each stored as a flight test data base (FTDB) along test point. Typically, 30 to 35 data points would with the raw engineering units from which it was be accomplished on a stability and control icing derived.
research flight, and the time required to complete these points would be approximately 30 to 40 min. Aircraft performance was determined by an On some initial icing research flights, both per- analysis program that took the thrust model of formance and stability and control tests were propulsion system, then accessed the FTDB, and accomplished; however, this procedure was discon- calculated lift and drag coefficients by means of tinued in favor of flights dedicated to perform- a basic set of equilibrium equations.
ance only or stability and control only. Combining performance and stability and control resulted in The MMLE analysis for the determination of a less rigorous investigation of each phenomenon aircraft stability derivatives was somewhat more due to the limitation of time brought about by complex. The MMLE method is based on an assumed ice shape deterioration. mathematical model of the airplane where the sta- bility and control derivatives represent the Data Analysis unknown parameters. Initial conditions and dynamic control inputs measured in flight were A brief summary of data handling and analysis applied to the model with starting values of the methods are given in this section of .the report. unknown parameters, and the complete response of The discussion is broken down into static and the model was then compared to that of the air- dynamic methods. No special modifications were plane. The difference was a response error. The made to the basic procedures in either method for MMLE program then changed the aerodynamic deriva- the artificial or natural icing flight tests. tives by a computational algorithm to reduce the response error. The new derivatives were then Static~itudin~l.light Tests with Artificial used in the math model to compute a new response Ice Shapes error. This iteration procedure continued until a specified convergence criterion was met. The Aircraft test weights and c.g. 10cativilS were final derivatives represent airplane aerodynamic determined for each flight maneuver by applying characteristics that minimize the error between fuel weights and moments to the zero fuel condi- airplane and mathematical model responses in the tion of the aircraft. Recorded flight parameters least squares sense.
that included true airspeed, angle of attack, pressure altitude, static air temperature, eleva- Power effects on the stability derivatives tor position, pilot elevator force, and vertical were handled by establishing baseline derivatives acceleration were calibrated, scaled, then tabu- as a function of flap setting and thrust coeffi- lated for each stabilized test point. Aircraft cient. Thus when a stability and control maneu- lift coefficient CL and normal force coefficient ver was performed with ice on the aircraft, the C corresponding to each of these stable points degradation in a particular derivative would be N A determined relative to the baseline derivative c1 imbs and were then calculated. For sawtooth calculated at the same power setting. All sta- bility and control analyses from MMLE contained versus CL descents, &e versus CL and Fe/q For wind-up were plotted for each c.g. tested. in this report are referenced to thrust coeffici- versus C ent.
turns, &e versus C and F / N e q NA A Results and Discussion were likewise plotted for each c.g. tested.
Curves for each of these parameters were hand Longitudinal Characteristics with Artificial Ice faired and, by the methods in Ref. 9, the Shapes Applied to the Horizontal Tail location of the stick-fixed and stick-free neutral and maneuver points were determined.
Control position and control force gradients from sawtooth climbs/descents and wind-up turns Transient Flight Tests for Performance and were examined for all five research flights. Each Stability and Control flight employing an artificial ice shape was com- Flight data from appropriate dynamic pared to the baseline flight. Inspection of the plots indicated that only subtle differences maneuvers were reduced and analyzed on a Gould existed between the baseline gradient data and SEL 32/77 computer with a dual processor located ."
gradient data from flights with surface roughness at KSR in Lawrence, Kansas. A block diagram of and the artificial rime shape. However, control the data management system is shown in Fig. 7.
force and control position gradients from the This data management system was designed to light and moderate glaze shapes displayed more function with the KSR Data Acquisition System.
recognizable trends away from the baseline. As The initial phase of the processing involved would be expected, the moderate glaze ice shape provided the greater deviation of the two.
encoding the flight tape to a SEL format, converting Qualitatively, the aircraft was always strong Flight 86-20 represented a moderate to heavy mixed positive stable during longitudinal maneuvers with iCing condition and 86-21 represented a moderate each artificial ice shape. However, the moderate glaze icing condition. Flights 86-16 and 86-17 glaze ice shape did cause noticeably lighter pitch represented mixed icing conditions and were selec- forces especially at lower speed. This observa- ted for longitudinal stability analysis. Flight tion was supported by the data; and thus, it was 86-16 provided a comparison bewteen the baseline, decided to calculate neutral and maneuver point all iced, then wing only iced condition, while locations for the moderate glaze ice shape and flight 86-17 compared the baseline, all iced, compare them to baseline data. These calculations then tail only iced condition. When looking at were not made for the other artificial shapes.
any flight data that shows "wings deiced," it must be remembered that the portions of the wing Figures 8, 9, 10, and 11, which summarize between the engine nacelles and fuselage are not the variations in neutral and maneuver point loca- protected; and, hence, approximately 17 percent tions with Cl, provide comparisons between the of the wing span cannot be deiced. The full baseline and moderate glaze ice shape. Figure 8 effects of this on performance and stability and shows that on the average, approximately a 7 to 8 control measurements cannot be assessed·at this percent reduction in static margin occurs for the time. It is anticipated that the lateral- stick-fixed case. For the stick-free case, this directional data will be presented in a follow- reduction averages 3 to 5 percent as shown in on report.
Fig. 9.
Icing cloud data. - Icing cloud data recorded Data from wind-up turns also showed a reduc- during natural icing flight tests are shown in tion in maneuver margln. Figure 10, for example, Figs. 13 to 17. The ,measurement of lWC, MVD, and shows that the stick-fixed maneuver point moves temperature are subject to the following consider- forward approximately 3 percent at the lower Cl'S ations: and as much as 12 to 15 percent at Cl'S near stall. For the stick-free case shown in Fig. 11, a. lWC and MVD are not corrected for local the maneuver point moves forward approximately droplet concentrations due to the presence of the 15 percent at Cl'S around 0.9, and 7 percent or aircraft in the icing cloud. A program is cur- less near stall. rently underway to calculate local droplet tra- jectories.
Generally, the data showed fairly consistent trends. Static and maneuvering margins were b. The Rosemount Ice Detector is an accretion- affected over the entire range of Cl'S tested type device and, therefore, calculates an average with the moderate glaze ice shape attached to the lWC after a small amount of ice accumulates on the tail. These results are consistent with those sensing portion of the probe. Rosemount Ice reported in Ref. 10. Detector plots of lWC tend to be somewhat jagged due to the coarse cycle time of the instrument, It should be pointed out that the static methods just described have many potential sources c. Discontinuities in FSSP data generally for error. Small variations in the measured para- occur for one of two reasons. Either the aircraft meters can be easily masked by instrument error, flys out of the cloud, or some component of the manual curve fitting, slope estimates, and extra- probe ices over and the instrument no longer func- polation techniques. Because of this situation, tions. Portions of some FSSP plots show areas of it was possible to report with confidence only missing data, which can be explained by knowledge those results where the effects of icing were of the flight profile through the icing conditions.
very pronounced. For example, Fig. 12 shows the relative differences between normalized force d. Sharp spikes in temperature profiles are gradient curves measured on the baseline tail and "noise," such as those shown in Fig. 14. The the tail with the moderate glaze ice shape. These jagged "sawtooth" dew point temperature profile differences were of sufficient magnitude to report as seen in Fig. 16 is an instrument problem.
static margin degradation with confidence. On flights with surface roughness, rime, and light A summary of these data, which are time glaze shapes, the differences were more subtle averaged, appear in Table III. This summary pro- and relative changes were hard to break out. It vides a quick look at the overall icing conditions should be noted again that these tests were con- encountered and can be referenced to the icing ducted with flaps up. Had the flaps been exten- certification requirements as contained in Federal l ded, the relative reduction in static margin Air Regulation (FAR) Part 25, Appendix C.1 would probably have been greater as described in Ref. 10, and the effects of the lesser ice shapes Ice shape documentation. - In-flight photo- more pronounced.
graphs and stereo analyses of wing ice shapes are provided in Figs. 18 to 22. As in the icing cloud Aircraft Performance and Stability and Control data, each figure is self-explanatory. The shape After Flight Through Natural Icing Conditions of the ice is critically important with respect to the performance degradation and stability and General. - Ten icing research flights were control effects on an aircraft. The "double-horned" flown in areas of northeastern Ohio, western ice shapes characteristic of glaze icing generally " Pennsylvania, and northwestern New York during cause the greatest penalties in lift and drag.
the month of December 1985. Eight of these flights Temperature is one of the primary factors that yielded research-quality data; and of these eight, influence the type of ice forming on a given air- four research flights were analyzed for this craft surface. Generally, glaze ice forms at the report. Discussions of results are referenced to warmer temperatures as shown by the icing and assigned flight numbers. Flights 86-20 and 86-21 photographic data for flight 86-21 (Fig. 19), were selected for a full perfurmance analysis.
while colder temperatures cause mixed accretions (Flights 86-20, 86-16, and 86-17 shown in Figs. 18, wing, and horizontal tail. The differences in lift 20, and 21 respectively), and rime accretions and drag coefficients at the two power settings (Flight 86-23 shown in Fig. 22). Notice that are essentially a maximum power effect. What is mixed ice accretion shapes tend to be somewhat also interesting is that lift and drag decrements, rectangular in cross section with small, longi- which were broken out by selectively deicing the tudinally oriented, finger-like protrusions at aircraft, remain relatively the same percentage the upper and lower edges, while rime ice takes wise at both high and low power settings. For on a more wedge-shapped appearance. Photographs example, referring to Fig. 23 and the summary per- taken on Flight 86-23 were included only to pro- formance data in Table IV for flight 86-20, the vide an example of rime ice. Stability and con- total degradation in lift coefficient at an arbi- trol analyses for this flight are forthcoming in trary angle of attack of 6' i; approximately a future report.
7 percent with 40 psi torque, and 8 percent with 15 psi torque; and the individual contributions of Aircraft performance. - Aircraft performance each iced component are nearly the same at both measurements were obtained for the baseline (non- power settings. The drag polars in Fig. 24 along iced) condition with flaps set at 0', 10', 20', with summary data in Table IV for the same flight and 37.5'; however, only the 0' flap data was show similar results. Percentage wise, the drag actually carried beyond the flight test data base decrements were about the same at the high and low level and plotted in coefficient form. It was power settings. These observations also hold for decided that only iced aircraft performance in performance data seen in Figs. 25 and 26. This is the 0' flap cruise configuration would be analyzed Significant because the relative magnitude of lift for this report.
and drag decrements due to ice appear independent of power within the linear portion of the flight Aircraft performance is presented in terms envelope. On the other hand, the absolute magni- of lift slopes and drag polars and may be found tude of these decrements are dependent on power in Figs. 23 to 26. A pitching moment curve, which effects; and it would seem that these observations is also derived from level accelerations and would hold for any conventionally configured, multi- decelerations and an MMLE analysis, is provided engine, propeller aircraft. This will become an in Fig. 27. Because of the rather substantial important consideration in the development and effect power has on these data, all icing flight validation of aircraft performance prediction tests were conducted at two basic power settings codes, which must take into account power effects.
and referenced to the respective baseline plot.
Level accelerations were performed at 40 psi The flight data point out that the shape of engine torque and 96 percent propeller RPM; and aircraft ice accretions on both lifting and non- level decelerations at 15 psi engine torque at lifting surfaces is the most important factor the same RPM. These power settings approximate influencing performance. For example, the icing maximum and minimum thrust conditions, and, there- encounter on flight 86-2~ lasted 54 min at an by, provide the full range of power effects on average LWC of 0.46 gm/m while the encounter aerodynamic performance coefficients. Baseline on flight 86-21 lasted j5 min at an approximate and icing flight data are then compared at these average LWC of 0.2 gm/m. Though the average power settings, and the absolute magnitudes of MVD on each flight was approximately 14 to 15 ~m, aircraft performance degradation due to the aero- the temperature differed by 4.5 'C. The differ- dynamic effects of icing are readily determined.
ence in temperature resulted in a glaze-type ice In this regard, the method of level accelerations formation on flight 86-21. The encounter on this and decelerations has an advantage over the steady flight was 9 min shorter than on the previous power for level flight technique since the latter flight 86-20 and the LWC less by a factor slightly technique incorporates a power effect in each greater than two; yet the overall drag increase discrete measurement, which is more difficult to was about 15 to 20 percent higher. The lift account for. curves don't show as large a variation from one flight to the other; however, it must be remem- Transients during the initial phase of each bered that CL measurements made at a 6' angle acceleration or deceleration maneuver are gener- of attack are still below stall buffet speeds.
ally disregarded in the analysis. Normally, this Previous experience and data appearing in Refs. 2 is not too much of a problem; however, when the and 3 show that the lift curve for the glaze ice aircraft is iced up, the speed envelope is smaller condition flattens more rapidly than those for due to a higher stall speed and a lower maximum other ice shapes at higher angles of attack.
speed. As a result, a relatively smaller range of coefficient values are calculated. This effect Longitudinal Stability and control charac- is seen throughout the performance analysis and teristics with ice. - Referring now to the identifed by the boundary labeled "ice data range."
pitching moment curves in Fig. 27, it becomes This boundary reflects the average level flight imnediately apparent that power effects literally speed envelope attainable for the type of icing overwhelm any changes in the static pitching encountered on a particular flight less that which moment parameter due to icing. Note the differ- was "clipped" during the initial transient •• Note ence in baseline slopes at 15 psi torque as that the drag polar data is extrapolated to a opposed to the 40 psi torque curve. The high CL = O. thrust line of the engines in combination with propeller slipstream effects over the wing and All the performance plot; show that lift tail greatly increase the static longitudinal curves and drag polars are steeper and of higher stability parameter C as shown in Fig. 27(a).
m absolute value at 40 psi torque than at 15 psi a torque. This is an expected occurrence as the At the 15 psi torque setting, the effect of icing propeller wake energizes the wing boundary layer on this parameter was not seen in the data from to increase lift and also increases drag due to either flight 86-20 or 86-21. At the 40 psi torque the wash of the slipstream over the engine nacelle, pressure setting, small reductions were measured in C on both icing flights. However, the data or 86-17. However, when flaps were lowered 10', m a a 25 percent degradation was seen for both flights from both flights was indistinguishable; that is, in any iced configuration. The data from both for the same iced configuration, the "mixed" type flights are shown on Fig. 33.
ice shape and the "glaze" type ice shape gave the same result. It was for this reason that the Conclusions data from both flights 86-20 and 86-21 were lumped onto the single plot.
The performance of an aircraft after flight through measured natural icing conditions is pri- Comparing baseline and iced slopes of C m marily affected by the shape of ice forming on versus a, the static pitching moment of the forward-facing airframe components. Performance aircraft is reduced approximately 13 percent in modeling methods, combined with dynamic maneuvers, the all-iced, power-on case and approximately was found to be a very practical and expeditious 9.9 percent after the wings are deiced. (Remember way of measuring lift, drag, and pitching moment that the inboard portions between the engine decrements in natural icing conditions. The nacelles and fuselage are not protected.) This acceleration and deceleration maneuvers enabled seems to be about the right order of magnitude the breakout of power effects, which in an abso- since static longitudinal tests with artificial lute sense, were determined to be very substantial ice showed an average 5 percent reduction in for a propeller-driven aircraft. A corollary to stick-fixed static margin at a lower power set- this fact was the observation that performance ting. Since the static pitching moment parameter decrernents were, percentage-wise, the same, in a is proportional to the static margin, these relative sense, over a wide range of power set- results appear reasonable.
tings. This was a very important finding with respect to its impact on the validation of per- Icing effects on longitudinal stability formance degradation codes and to other applica- derivatives for the baseline aircraft were tions such as the development of flight simulator compared with those obtained after a natural software for icing scenarios.
icing encounter.
Ice accretions (real and artifical) degraded Flights 86-16 and 86-17 were flown in mixed static and maneuver margins, elevator control conditions, and the ice shapes which formed on derivatives, and pitching moment slopes throughout the aircraft during each flight were similar.
the flight envelope. Elevator control derivatives This was an extremely fortunate circumstance as were more affected at higher tail downwash angles these similarities allowed a comparative analysis resulting from wing flap extension.
of the two flights. On flight 86-16 the effect of ice was studied for the all-iced and tail-only- The use of the KSR systern to acquire per- iced configurations at both zero and ten degree formance and stability and control data was found flap settings. On flight 86-17 the effect of ice to be extremely useful for natural icing flight was also studied on the all-iced aircraft; how- tests. The rapidity with which data can be ever, this time the tail was deiced and the effects acquired by transient response techniques makes of wing ice alone were evaluated at the zero and this systern a viable method in a time-critical ten degree flap settings. Looking first at the test environment.
zero degree flap results for fligh~s 86-~6 and 86-17 in Figs. 28 and 30, the all-lced alrcraft One of the long range objectives of the NASA showed an average 10 percent degradation in eleva- Lewis icing research program is to provide method- tor power C ,and elevator effectiveness, C L ologies that can predict aircraft performance m6 6 e e losses and stability and control effects on an When the tail only was iced (Fig. 28), this aircraft. This technology would have direct degradation averaged around 8.5 percent; and when application in: aircraft design; conducting the wing only was iced (Fig. 30), a rather small sensitivity studies for advanced flight control 2 percent degradation was calculated. When flaps systems on relaxed stability aircraft; performing were lowered ten degrees, the all-iced degradation failure effects modes analyses for ice protection in these coefficients increased to a more signifi- systems; generating simulator software for pilot cant 15 to 16 percent as shown in Figs. 29 and 31. training; and in improving aircraft icing certi- It was interesting that wing ice, which had little fication criteria for better operational safety.
effect on the derivatives with flaps up, now caused The test results reported herein are an initial a degradation in elevator power and elevator step in that direction. Advancements in this effectiveness of approximately 9 percent. technology will require a more rigorous approach to the problem.
Fi9ure 32 shows the effects of ice on a com- bined pitch damping state coefficient whose indi- References vidual componets were not identified by MMLE. It was found in the flaps-up case, icing had no effect 1. Bowden, D.T., Gensemer, A.E., and Skeen, C.A., on this coefficient; however, when flaps were "Engineering Summary of Airfran:e Icing .
lowered 10' and ice was on the wings, tail, or Technical Data," General Dynamlcs/Convalr, both, this coefficient was degraded approximately San Diego, CA, Technical Report ADS-4, Mar.
'.
23 percent. Data from flights 86-16 and 86-17 1964.
provided identical results and were, therefore, plotted on the one figure. A similar result was 2. Ranaudo, R.J., Mikkelsen, K.L., McKnight, seen in Fig. 33 for the combined state coefficient R.C., and Perkins, P.J., Jr., "Performance (C + C .). With flaps up, no degradation was Degradation of a Typical Twin Engine Commuter L L q a Type Aircraft in Measured Natural Icing seen in this coefficient for either flight 86-16 Conditions," NASA TM-83564, 1984.
3. Mikkelsen, K.L., McKnight, R.C., Ranaudo, 8. Crowley, L.D., "Trailing Cone Systems Applications," Douglas Aircraft Company; R.J., and Perkins, P.J., Jr., "Icing Flight TM-GEN-4158, Aug. 1967.
Research: Aerodynamic Effects of Ice and Ice Shape Documentation with Stereo Photography~" 9. Perkins, C.D., "Static Longitudinal Stability NASA TM-86906, 1985.
and Control," pp. 3:1-3:29, and Johnson, 4. Shaw, J.J., "Progress Toward the Development H.I., "Flight Testing Aircraft for of an Aircraft Icing Analysis Capability," Longitudinal Maneuvering Characteristics," NASA TM-83562, 1984. pp. 4:1 - 4:31; AGARD Flight Test Manual, Vol. II, Stability and Control, edited by 5. Renz, R.L. and SChweiknard, W.G., "A Miminum C.D. Perkins, 2nd revised ed.; Pergamon Approach to Flight Testing," Flight Testing - Press, Oxford, 1963.
Evolution and Revolution, Society of Flight Test Englneers, 1985. 10. Ingelman-Sundberg, M. and Trunov, O.K., "On the Problem of Horizontal Tail Stall Due to 6. Ide, R.F. and Richter, G.P., "Evaluation of Ice," Swedish Soviet Working Group on Scientific-Technical Cooperation in the Field Icin9 Cloud Instruments for 1982-83 Icing Season Flight Program," AIAA Paper 84-0020, of Flight Safety, Report JR-3, Feb. 1985.
Jan. 1984.
11. "Ice Protection," Airworthiness Standards: 7. McKnight, R.C., Palko, R.L., and Humes, R., Transport Category Airplanes, F.A.A.
"In-Flight Photogrammetric Measurement of Regulations, Part 25, Section 25.1419, and Wing Ice Accretions," AlAA Paper 86-0483, Appendix C, 1974.
Jan. 1986.
TABLE I. - AIRCRAFT CHARACTERISTICS Wing area, ft2 420 Wing span, ft . 65 Aspect ratio • 10 6.5 MAC, ft •.•. ~ Aileron area, ft . 14.68 Elevator area, ~t2 21.61 Rudder area, ft . 35.55 11 000 Standard weight, lb Moments of inertia at standard weight during icing flight test period: lxx, slug ft2 16 237 I , slug ft2 23 433 yy I , slug ft2 36 312 zz I , slug ft2 1 141 xz TABLE II. - KSR DATA ACQUISITION SYSTEM [Recorded Data, NASA Twin Otter DHC-6-004.]
Variable Variable Variable name Units number 1 FILCNT File counter 2 BLKCNT Block counter 3 ESB Engineer's status BYTE Aircraft status BYTE 4 ASB Pause event Rosemount heat Gear position Gyro erection marker Pilot event sec 5 Time Time Longitudinal acceleration G 6 AX 7 G AY Lateral acceleration G 8 AZ Z-direction acceleration 9 Pitch-rate Pitch rate deg/sec 10 Roll-rate Pitch rate deg/sec 11 deg/sec Yaw-rate Yaw rate Pitch att itude deg 12 Pitch-att 13 Roll-att Ro 11 att itude deg psf 14 Delp-alpha Pressure alpha 15 Delp-beta Pressure beta psf Pressure reference psf 16 Delp-ref 17 deg Delta-A-L Aileron deflection empty 19 deg Delta-E Elevator deflection deg 20 Delta-R Rudder deflection 21 deg Flap Flap position psf 32 Diff-press Differential pressure 33 Air-temp Indicated total deg/K temperature 34 Inert-vref Vertical gyro reference volt voltage 35 Stat-press Static pressure psf Reference voltage volt 36 Cpt-v ref 37 Battery-A Reference battery board 1 volt lb 39 Fuel-used Fuel used 40 PAF Pilot aileron force lb 41 PEF Pilot elevator force lb 42 PRF-L Pilot rudder force, left lb Pilot rudder force, right lb 43 PRF-R 49 N1-L Gas generator RPM-N1, left percent Gas generator RPM-N1, right percent 50 N1-R 51 PROP-RPM-L Propeller RPM, left percent percent 52 PROP-RPM-R Propeller RPM, right 53 psi Torque-L Engine torque pressure, left Torque-R Engine torque pressure, 54 psi right lb/hr Fuel-Flo-L Indicated fuel flow left 55 engine lb/hr 56 Fuel-Flo-R Indicated fuel flow right engine 57 deg/K Fueltemp-L Fuel temperature left engine - deg/K 58 Fueltemp-R Fuel temperature right engine 64 Liquid water content volt Johns-will indicator 65 Leigh Ice detector unit volt Ice detector unit volt 66 Rosemont 67 Gen-East Dew point hygrometer volt Reference battery board 2 volt 71 Battery-B TABLE III. - SUMMARY OF AVERAGED ICING CLOUD DATA FOR PERFORMANCE AND STABILITY AND CONTROL TAS, AOA, Static Dew Average Average Flight Start End Alt. , point, MVD, time PA, Kts deg temperature, number time LW~~ gom m °c °c ft 11 ----- 0.25 19 14:32:58 15:12:18 8060 138.6 1.4 -8.0 - 8.1 .33 10:08:28 10:27:18 7309 135.0 1.6 -7.2 .46 14 127.3 1.8 -9.5 -10.0 20 12:42:48 13:36:58 6163 130.8 1.6 -5.0 - 4.8 -.20 15 21 09:55:38 10:40:38 4315 -0.5 -10.7 .30 10 23 10:13:38 11: 11:48 4330 136.4 0.5 TABLE IV. - SUMMARY OF ESTIMATED PERFORMANCE LOSSES DUE TO ICING Flight Airframe component(s) Ice isolated Parameter Total effect on Component contribution number type by selective deicing parameter, to total effect, % % Engine torque 40 psi 15 psi 40 psi 15 psi 86-20 Mixed All components -8 -- -- -7 Wings 36 40 Tail, wing strut, and main 36 27 gear strut Miscellaneous components CL 28 33 (flap hinges, antenna, etc.) at 86-21 glaze All components (a = 6°) -8 -7 Wing 47 43 Tail, wing strut, and main 24 21 gear strut Miscellaneous components 29 36 86-20 Mixed All components +36 +30 Wing 36 40 Horizontal stabilizer 26 27 Vertical stabilizer wing strut 23 20 and main gear strut Miscellaneous components CD 13 86-21 Glaze All components +50 +47 at Wing (CL = 0.5) 30 28 Vertical stabilizer 16 17 Horizontal stabilizer 9 13 '.
Wing strut and main gear strut 16 17 Miscellaneous components DEW POINT I ROSEMOUNT TEMPERATURE "\ / ICE DETECTOR / \ / \ /
E§
\ I \ OAp..J '- FSSP ROSEMOUNT TOTAL AIR TEMPERATURE PROBE---- ROSEMOUNT ICE DETECTOR-- ._--- AFT STEREO CAMERA PORT \" VERTICAL TAIL \ DE-ICER BOOT \ \ {- ROSEMOUNT \ ' I 858 PROBE I I I I I @ --"'-- WING AND LANDING GEAR PNEUMATIC DE-ICER BOOTS Figure 1. - NASA Lewis Research Center icing research aircraft and locations of icing instruments.
(a) Liquid water content instruments (a) Johnson Williams, (b) Rosemount ice detector, (c) pressure ice rate and accretion meter, (d) Leigh ice ,detector._ (e) CS I RO-K I NG.
(b) Droplet sizing instruments (a) forward scattering spectrometer probe, (b) optical array probe.
Figure 2. - Icing instrumentation.
AFT CAMERA., FLASH I rFORWARD I HEAD, \ / CAMERA I \ I I \ I I I WAKE SURVEY ...- PR OBE------ Figure 3. - Wing stereo camera system layout.
(a) Pi lot's dynamic force wheel.
(b) Rudder pedal load cells.
Figure 4. - Dynamic force wheel and rudder pedal load cells for preform- ance and stabi lity and control measu rements.
Figure 5. - Heated Rosemount 858 flow angle sensor to measure airspeed, altitude, angle of attack, and sideslip.
(a) Artificial "Ri me"ice shape.
(b) Artificial "Light Glaze"ice shape.
Figure 6. - J\rtificial ice shapes attached to the horizontal tai I plane.
(c) Artificial "Moderate Glaze"ice shape.
(d) Full span view of the moderate glaze ice shape.
Figure 6. - Concluded.
ENCODE KSR DATA I------..a FORMA TTIN G 1--------1 .....
COUNTS CALIBRATION AIC DATA & INPUT MANUAL COUNTS TO INSTRUMENT CAL DATA ENGINEERING CALS UNITS 2nd ITERATION TEST TIME I---~t GUIDE HISTORIES PLOTTER CONVENTIONAL STATIC TAKEOFF STABILITY PERFORMANCE PERFORMANCE STABILITY LANDING DERIVATIVES CLiMBI ACCELI MODELING LONG. LAT.
GROUND MMLE DIR.
CRUISE EFFECT PROCESSED PLOT DATA BASE PROGRAM Figure 7. - KSR data management system.
-I-' C Q.)
~ 60 Q.)
0- 0...
I--~
"
~
',0
o a.. .....
, -l 50 « n::: '0, I-- , =:l lLJ Z -0-- BASELINE (NO ICE)"«..
Cl --<)-- WITH ARTIFICIAL , ~ 40 w:: ICE SHAPE I ~ u i= V'l 30~ ____ ~ ______ ~ _____ ~ ______ ~ ____ -l .8 .2 .4 .6 1.0 1.2 Figure 8. - Change in stick-fixed neutral pOints with artificial ice on horizontal tail. Nominal cruise power, 275 SHP per engine.
+-' C ~ 60 I- Q.)
0- I--~
- O
~ -- -~
O-_ ....... Q
o .......
~ 50 ......
« n::: I-- ---0--- BASELINE (NO ICE) '«..
=:l lLJ --0-- WITH ARTIFICIAL Z ICE SHAPE t:!::l 40 n::: I.J...
I ~ u i= V'l
30.2~----~----l~--~----~~~
1.2 Figure 9. - Change in stick-free neutral points with artificial ice on horizontal tail. Nominal cruise power, 275 SHP per engine.
u 70 « --0- BASELINE 2: .......
--0-- WITH ARTIFICIAL c: (]) ICE SHAPE , U I- , ~ 60 , I-~ , ~ , o 0- , ,
8J 50
> ::::>
0'
I..L.I \ z « \ 2: \
8 40
x
u: ~
I ::..::: u i= ~ 30~----~----~------~----~ .6 .8 1.2 1.4 Figure 10. - Change in stick-fixed maneuver points with artificial ice on horizontal tail.
Nominal cruise power, 275 SHP per engine.
V CAL ::: 103 knots.
BASELINE ---0-- WITH ARTIFICIAL --0-- u 70 ICE SHAPE « :2: .......
c: (]) U I-
\ ~ 60
\ e::: I..L.I
~ ~
0- \ e:::
\
I..L.I > \ ::::> I..L.I
\
z « \ 2: I..L.I 40 0\ I..L.I e::: \ I...L.
I \.
::..::: , u ......
i= 30 ~ .6 .8 1.0 1.2 C L Figure 11. - Change in stick-free maneuver points with artificial ice on horizontal tail. Nominal cruise power, 275 SHP per engine. V CAL::: 103 knots.
-1. 6 STANDARD WEIGHT = 11 000 Ib c. g. :: 32 ± 1 percent MAC -1. 2 .Q"' (l) BASELINE "" u..
\ , LJ..r PULL u -.8 0::: I , ~ u ',- MODERATE GLAZE -.4 i= V> I.J.J t:::!
-l
a
<C ::2: 0::: z PUSH .4 .8
a .2 .4 .6 .8 1.0 1.2 1.4
Figure 12. - Normalized longitudinal control force variation with lift coefficient for the clean horizontal tail versus the moderate glaze ice shape attached to the horizontal tail (OF = 0°)'
1.2
ROSEMOUNT ICE DETECTOR C""\E .8 O"l
-
c...i ~ .4 FSSP MISSING
I' MISSING DATA
'I ~A
1y'V"/\~
~ ,J\.\!t" J~
, \
J,hl I' v' \ 10' I' \
: \
'I \ ~, I
0: ~ I I
:~: ); u 1..1..f e::: :::> -10 I- « e::: " DEW POINT \.J.J 0- o STATIC -20 :.?: \.J.J o STATIC I- -30 0 10 20 30 40 50 60 TIME, sec Figure 13. - Icing data for performance flight 86-20.
.6 (\'\ ROSEMOUNT ICE DETECTOR
: .3 .f\ ~"' . _ ~ ~
O~~~~-~
MISSING FSSP 30 DATA
H
u LW~ e::: :::::> ~10 I- A. DEW POINT « e::: o STATIC LW ~20 CL.
o STATIC ~ LW I- -30 0 5 10 15 20 25 30 35 40 45 TIME, min Figure 14. - Icing data for performance flight 86-21.
.6 ROSEMOUNT ICE DETECTOR ("("\ E cr, u .3 ~ FSSP E :::1. 20 Cl ~ u o LJ.J~ 0::: => I- « Do DEW POINT (INOPERATIVE) 0::: LJ.J o STATIC 0.. -20 :2: o STATIC LJ.J I- -30L----L---~--~----~--~--~--~--~--~
o 5 10 15 20 25 30 35 40 45
TIME, min Figure 15. - Icing data for stability and control flight 86-16.
.8 ROSEMOUNT ICE DETECTOR ('t'I E O'l
-
.4 u~ $ -J FSSP
I- MISSING DATA
E ::1.
~\ (\v\j~~
I \ I I I \...JIJ I I > I 2: I 10 I I u o L.L.t §3 -10 I- « 0::: A DEW POINT ~ -20 o STATIC 2: LJ.J o STATIC I- -30~------~--------~------~~ _______ ~
o
TIME, min Figure 16. - Icing data for stability and control flight 86-17.
ROSEMOUNT ICE DETECTOR .6 ('/"\ E I::n .3 u $: ....J FSSP E :::1.
~ 10 I I I
!5' 0
A DEW POINT STATIC STATIC
I I I
I I I
-30
o 20 30
10 40 50 60 TIME, min Figure 17. - Icing data for stability and control flight 86-23.
3r- "GLOBULE" FORMATION SEPARATE FROM MAIN ACCUMULATION 7 / / / / 21-
It'
/~
(
i+
~ ~
-u-
fi5> g>,~ '-S2 ~--cR9@cP -21- -3! I I -2 -1 0 THICKNESS, in.
Figure 18. - Wing stereo analysis and photograph of wing strut for performance flight 86-20 for moderate to heavy mixed icing conditions.
VIEW OF ATTACHMENT POINT WAS BLOCKED 7 C 1-- :::r: 0 <..?
L.lJ :::r: -1 -2 -3 I I ! I -2 -1 0 1 .
THICKNESS, in.
Figure 19. - Wing stereo analysis and photograph of empennage for performance flight 86-21 for moderate glaze icing conditions.
STEREO ANALYSIS NOT AVAILABLE Figure 20. - Wing ice shape and empennage photographs for stability and control flight 86-16 for mixed icing conditions.
c
1-- :r: <:) i:iJ :r: -1 -2 -3 I I f I J -2 -1 0 1 2 THICKNESS, in.
Figure 21. - Wing stereo analysis and photograph of empennage for stability and control flight 86-17 for mixed icing conditions.
Figure 22. - Ri me ice on empennage duri ng flight 86-23.
1.3 BASELINE WING, TAIL, AND STRUTS DEICED WING DEICED 1.1 ALL ICED .9 .7 .5 d .3 (a) Acceleration maneuvers with 40 psi torque (OF = 0°).
1.1 .9 .7 .5 .3~~~----~----~--~----~--~ -2 0 2 4 6 8 10 ANGLE OF ATIACK, a, deg (b) Deceleration maneuvers with 15 psi torque (OF = 00).
Figure 23. - Effect of mixed ice on lift curve for perform- ance flight 86-20.
ALL ICED WING DEICED WING AND HORIZ. STAB.
DEICED WING, VERT, AND HORIZ.
STAB., AND STRUTS DEICED BASELINE c u .12 (b) Deceleration maneuvers with 15 psi torque (~ = 00).
Figure 24. - Effect of mixed ice on aircraft drag for performance fl ight 86-20.
1.3 BASELINE WING, TAIL, AND STRUTS DEICED.
WING DEICED 1.1 ALL ICED .9 .7 .5 ...J U 1.3 1.1 .9 .7 .5 .3~~~ ____ ~ ____ ~ ____ ~ __ ~
-2 o 2 4 6
ANGLE OF ATIACK, a, deg (b) Deceleration maneuvers with 15 psi torque (6F = aD).
Figure 25. - Effect of glaze ice on I itt curve for performance flight 86-21.
---- ALL ICED --- WING DEICED -- WING AND VERT. STAB.
DEICED ---- WING, VERT. AND HORIZ.
STAB. DEICED ---- WING, VERT. AND HORIZ.
STAB .• AND STRUTS DEICED .14 BASELINE .12 .10 .08 .06 .M~~~~~--~--~~~~~ (a) Acceleration maneuvers with 40 psi torque c!f (OF = 0°1. .
. 14 .12 .10 .08 .06 .MO~--~--~--~--~--~~~ .2 (b) Deceleration maneuvers with 15 psi torque (OF = 0°).
Figure 26. - Effect of glaze ice on aircraft drag for performance flight 86-21.
---- ALL ICED --- WING DEICED - - WING, TAIL, AND STRUTS DEICED BASELINE .1
o
-.1 E u -.2 (a) Acceleration maneuvers with 40 psi torque (6F = 0°1.
.1
o
-.1 '- __ L--_---' __ --'- __ -L-_----'
-2 o 2 4 6 8
ANGLE OF ATIACK, 0, deg (b) Deceleration maneuvers with 15 psi torque (6F = 0°). Ice had no effect.
Figure 27. - Effect of ice on C versus o.
m Results were the same for performance 86-20 and 21.
.003 ALL ICED ~ I
-
C'I TAIL ICED ---1 I
Q) .002 "C
t
u .. i 9.50/0 5:2 Q) <0 E 8.00/0 .001 u <l C BASELINE = -0.0285 deg- mo
e
O~------------~------------~--------~~~----- (a) Degradation in elevator power derivative.
-.0010 ALL ICED I
-
Cl Q) "C TAIL ICED
....... ~
-.0005 5:2
T 11.60/0
Q)
::r
u CIf> BASELINE = 0.0069 def 8.70/0 <l e .10 .15 .20 .25 MACH NUMBER (b) Degradation in elevator effectiveness derivative.
Figure 28. - Effect of mixed icing conditions on the elevator derivatives for stabil ity and control flight 86-16 (OF = 00 and CT = 0.06 to 0.081.
.004 ALL ICED .003 I
-
Q.)
"t:I LLI TAIL ICED U .002 Q.)
cO E u 15.1 % <1 9.1 % .001
e BASELINE = -0.232 deg-
mo
e
o~------~------~------~~--~--~ (a) Degradation in elevator power derivative.
-.0010 ALL ICED .-< I Q.)
"t:I LLI" TAIL ICED U -.0005 Cl BASELINE = 0.0056 deg- 16. 1 0/0 8.9% o e
o .05 .10 .15 .20
MACH NUMBER (b) Degradation in elevator effectiveness derivative.
Figure 29. - Effect of mixed icing conditions on the elevator derivatives for stability and control flight 86-16 (OF = 100 and CT = 0.05 to 0.07).
.'
.003 All ICED ......
I O'l OJ ,002 "C
I
LLi u 9.5% OJ <0 WING ICED 'I E .001 I u I <J C4; BASElINE = \ I e -0. 0285 deg- (a) Degradation in elevator power derivative.
-.0010 ......
I All ICED ----, O'l OJ "C L1..r u -.0005 11.6 % WING ICED "I \ C BASELINE = \ mo e O. 0069 deg- O~------~------~-----L--~- .10 .15 .20 .25 MACH NUMBER (b) Degradation in elevator effectiveness derivative.
Figure 30. - Effect of mixed icing conditions on the elevator derivatives for stability and control flight 86-17 (OF = oD and C = 0.06 to 0.08).
T .004 ALL ICED .003 .....
I 0> Q) 'C UJ- u .002 WING ICED Q) <0 E u
T:l%
<:l .001 8.6 % C BASELINE = -0.0232 def mO e o~------~------~--------~~--~~ (a) Degradation in elevator power derivative.
-.0010 ALL ICED .....
I 0> Q) 'C WING ICED ~ -.0005 Q) .:2 CLf> BASELINE = 0.0056 def u 8. 9 % 16. 1 0/0 <l e
o .05 .10 .15
.20 MACH NUMBER (b) Degradation in elevator effectiveness derivative.
Figure 31. - Effect of mixed icing conditions on the elevator
derivatives for stability and control flight 86-17 (OF = 100
and C = 0.05 to 0.07).
T (C + C ) BASELINE = -48 rad- mq ma ~=oo l1(C + Cma) ICE = 0 mq (a) Degradation in pitch damping coefficient was indis- cernable at ~ = 00 and CT = 0.06 to 0.08.
ALL ICED} ....
I 6
TAIL ICED ---------r--
"C <0 ....
WING ICED I
u..i u .----......
'0 4 E u 23% + C7 E ~ <I 2
o .05 .10 .15 .20
MACH NUMBER (b) Degradation in pitch damping state coefficient at br = 10 and C = 0.05 to 0.07.
T Figure 32. - Effect of mixed icing conditions on pitch damp- ing state coefficient for flights 86-16 and 86-17. Note, results were the same for both flights.
(CLq + C ) BASELINE = 10.8 rad-
La
OF = 0
ll(CLq + C ) ICE = 0
La
(a) Degradation in state coefficient was indiscernable at OF = 00 AND CT = 0.06 to 0.08.
-2 .....
I -0 ALL ICED } co ....
TAIL ICED ------r
Lo..i
WING ICED I
u
--:c- -1
-I u 25.40/0 +
o .05 .10 .15 .20
MACH NUMBER
(b) Degradation in state coefficient at Or = 10 and
CT = O. 05 to O. 07.
Figure 33. - Effect of mixed icing conditions on the state coefficient for flights 86-16 and 86-17. Note, results were the same for both flights.
3. Recipient's Catalog No.
2. Government Accession No.
1. Report No. NASA TM-87265
AIAA-86-9758
4. Title and Subtitle 5. Report Date
The Measurement of Aircraft Performance and
6. Performing Organization Code
Stability and Control After Flight Through
Natural Icing Conditions
505-68-11
7. Author(s) 8. Performing Organization Report No.
Richard J. Ranaudo, Kevin L. Mikkelsen, Robert C.
E-2962
McKnight, Robert F. Ide, Andrew L. Reehorst, '.
Jerry L. Jordan, William C. Sch1nstock, and 10. Work Unit No.
Stewart J. Platz
9. Performing Organization Name and Address 11. Contract or Grant No.
National Aeronautics and Space Administration
Lewis Research Center
Cleveland, Ohio 44135
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
15. Supplementary Notes Prepared for the 3rd Flight Testing Conference, cosponsored by the AlAA, AHS, CASI, DGLR, IES, ISA, ITEA, SETP, and SFTE, Las Vegas, Nevada, April 2-4, 1986. Richard J. Ranaudo, Kevin L. Mikkelsen, Robert C. McKnight, Robert F. Ide, and Andrew L. Reehorst, NASA Lewis Research center; Jerry L. Jordan, William C. Schinstock, and Stewart J. Platz, Kohlman Systems Research, Inc., Lawrence, Kansas 66044.
16. Abstract
The effects of airframe icing on the performance and stability and control of a
twin-engine commuter-class aircraft were measured by the NASA Lewis Research
Center. This work consisted of clear air tests with artificial ice shapes
attached to the horizontal tail, and natural icing flight tests in measured icing
clouds. The clear air tests employed static longitudinal flight test methods to
determine degradation in stability margins for four simulated ice shapes. The
natural icing flight tests employed a data acquisition system, ~h1ch was provided
under contract to NASA by Kohlman Systems Research Incorporated. This system
used a performance modeling method and modified maximum likelihood estimation
(MMLE) technique to determine aircraft performance degradation and stability and
control. Flight test results with artificial ice shapes showed that longitudi-
nal, stick-fixed, static margins are reduced on the order of 5 percent with flaps
up. Natural icing tests with the KSR system corroborated these results and
showed degradation in the elevator control derivatives on the order of 8 to
16 percent depending on wing flap configuration. Performance analyses showed
the individual contributions of major airframe components to the overall degrad-
ation in lift and drag.
17. Key Words (Suggested by Author(s)) 18. Distribution Statement
Aircraft icing; Aircraft performance; Unclassified - unlimited
Stability and control STAR Category 08
22. Price' 19. Security Classll. (of this report) 21. No. of pages 20. Security Classll. (of this page)
Unclassified Unclassified
-For sale by the National Technical Information Service, Springfield, Virginia 22161