Full report

Original page 1

3 1176 00168 0678 1 NASA Technical Memorandum 81707 NASA-TM-81707 19810010552 Survey of Aircraft Icing Simulation Test Facilities in North America William Olsen Lewis Research Center Cleveland, Ohio I February 1981 170_ REFEPL_,L'C,,. , " : ..7 .!!'!,:::i,.:::i

Original page 2

Original page 3

SURVEYOF AIRCRAFTICINGSIMULATIONTEST FACILITIESIN NORTHAMERICA William Olsen National Aeronauticsand Space Administration Lewis Research Center Cleveland,Ohio SUMMARY i A surveywas made of the aircraft icing simulationfacilitiesin North America. This was requestedof NASA by severalcommitteesconcernedwith Aircraft Icing. A similarsurvey of Europeanfacilitieshad already been reported in AGARD advisory report 127. There are 12 wind tunnels, 28 engine test facilities,6 aircrafttankers _ and 14 low velocityfacilities,thatcan performvarious aircraft icing tests _ full or part time. The surveydeterminedtl_elocationand size of the facility, its speed and temperaturerange, icingcloud parameters,and the technicalperson to contact. These resultsare presented in tabularform. The capabilitiesof each facility were estimatedby its technicalcontact person. The adequacyof these facilitiesfor various types of icing tests is discussed. lYe!l-19o

Original page 4

INTRODUCTION Aircraft that fly low and slow and have small aerodynamicallycritical surfaces are especiallysensitiveto icing. Helicoptersand generalaviation aircraft exactlyfit this worst case description. Significantactivity in the aircraft icingfield is expected because a large number of these aircraft are expectedto be developedto fly into icing conditionsin the next few decades. Considerabledata (bothR&D and Certification)is required before any aircraftcan be developedand certifiedas being able to fly safely through atomosphericicingconditions. Test flights in natural icingclouds are no great hardshipfor long range aircraft becausethey can ° rapidly fly the.greatdistancesand to the altitude required to find those elusive icing clouds. But obtainingflight data for the helicopterand GA aircraft in natural icing can be prohibitivelyexpensiveand time consuming becauseof their range and altitude limitations. In order to reduce our reliance upon natural icing flightsfor short range aircraft,improvements appear to be needed in the Aircraft Icing SimulationFacilitiesand in the analyticalmodels for icing and its effects. The first step in this improvementprocess is to determinethe capabilitiesand limitationsof all of the existing icing simulationfacilities. NASA was requestedto survey the capabilitiesof the facilitiesin North America that can do aircraft icing simulationtests. The survey was requested of NASA by the Standing Committeeon Icing, which is jointly sponsored by NASA, FAA and NOAA; similiarrequestshave also been made by the military servicesand AGARD. European icingfacilitieswere not included because they have already been surveyed (ref. i). The reasonsfor the survey are to: 1. Assist the icing researchcommunity in determingthe adequacyof the presentmix of icingtest facilities for all types of aircraft,2. Make it easier for a potentialfacility user to select and contact the icing facility that is appropriatefor his test requirements,and 3. Help facilitymanagers evaluate and improvetheir facility. This paperincludes a short descriptionof the various types of facilities, a detailed listingof the capabilitiesof each facility,and some discussionand evaluationof these capabilities. The capabilitiesof the facilitiesare presented in tabularform. The capabilitiesof each facility are the opinion of the tecnnicalpeople working with that facility. Based upon the informationin the tables and additionalinformation,cursory evaluationsare then made of the adequacyof the existing facilitiesfor the various types of icing tests. Some additionalcomments are also made about icing cloud instruments. DISCUSSIONOF FACILITIES The icingenvironmentthat an aircraft and its componentsmust operate in is describedin this section. Then the various types of icingfacilities,.andthe tasks they are used for, are briefly described. Following that, the capabilitiesof the icingfacilitiesin North America are briefly discussedalong with additionalinformation. ! Icing EnvironmentRequirements Aircraft flying throughclouds below about 8,000 meters can be subject tothe formationof ice (icing)on critical surfaces,which can cause serious losses in performanceand damage,and even a crash. The ice forms from 2

Original page 5

the small supercooleOdroplets in these clouds (IcingCloud Environment, ICE). At low altitudes,large supercooleddroplets (FreezingRain, FR) also result in icing. The effect of snow, and ice particlesand ice chunks on the aircraft (especiallythe engine)must also be considered. In addition, there is a concern about mixtures of the above conditions. The range of atmosphericparametersfor ICE, that are used for design and certification testing of transportaircraft,are defined in Federal AviationRegulation (FAR) part 25, appendixC (ref. 2). These ranges of temperature,liquid " water content and drop size, at various altitudesfor Stratiform(layer) and Cumuliformclouds, are shown as envelopeson figure 1. These envelopes define the maximum likely ranges of these parametersthat would Occur in nature (i.e., 9_.9 percent of the observationsin nature lie within these envelopes). The aircraftmanufacturermust design his aircraft to cope With every combinationof the parametersrepresentedby the envelopes,along with the mission of the aircraft (e.g.,altitude,airspeed and exposure time to the ICE). From these considerationshe must determinethe specific ICE conditionsthat result in the most severe icingon each aircraftcomponent (e.g.,wings, engine inlet,etc.). These discreteconditionsbecome the design and test conditionsfor icing tests of the aircraft,its components and icingprotectionsSystems. Although desirable,an icing facility doesn°t have to operateover the entire range of the entire FAR 25 envelopes in order to do meaningfulR&D tests and certificationtests. Furthermore, some aircraftcan't possibly encounterthe full range of conditions';indicated by figure 1. For example, the helicopterhas a limitedaltitude capabilitywhich makes high levels of LWC extremelyunlikely;indeed, reference 3 suggeststhat a truncationof the FAR 25 envelopesshould be used for helicopters. At the other extreme,engines are designed and tested for the whole FAR 25 envelope (ref. 4), largelybecausethey are used in a variety of aircraft. As a minimum goal, all facilitiesshould be able to produce 20 micron droplets for any icing test of a full scale aircraft or component. Types of Icing Facilitiesand Their Uses The types of icing test facilitiesand the types of icing tests are listed below. FACILITIESFOR ICING TESTS Natural icing flights Icing simulationfacilities: A. Wind tunnels B. Engine test facilities a. Free jet b. Direct connect C. Low velocityfacilities D. Flight tests with tankers Other icing simulationtechniques 3

Original page 6

TYPESOF ICINGTESTS Certificationand R&D testsfor: Engines Instruments Fixed wing aircraft Helicopters 0 Componentsof the above (including: ice protectionsystems, wings, etc.) General research and technology Natural icing flights.- Aircraft manufacturersoften fly their aircraft in a broad range of natural icingconditions in order to obtain certification by the FAA as being able to fly safely in icing conditions. This is an expensiveundertaking. It is reasonablefor long range aircraft,because they can fly greatdistances to fina those elusive icing clouds that are hard to find when you want them. But for short range aircraft (e.g., helicopters and civil aviation aircraft) a test program involvingnatural icing is all but prohibitivelyexpensive,time consuming and uncertain(refs. 5 and 6). In any event natural icing flight tests are not discussedin this paper, which is devoted entirely to Icing SimulationFacilities. Icing simulationfacilities.- Research and Developmentand Technology types of tests, and much certificationmust largely be accomplishedin one or more of the varied types of Icing SimulationFacilities. The four general types of simulationfacilitiesand their major variationsare schematically sketchedon figures 2(A) to (D). The primarydifferencesbetweeneach type of facility are in their geometry,airspeed,and in the types of tests they are used for. The general operationof all is similiar;the next paragraph describestheir operation in a generalway. In all icing simulationfacilities,the test aircraftor component is tested in a cold airstreamwhich contains a smaller icing cloud. The icing cloud is made up of either supercooleddroplets,which freeze when they strike the test surface,or ice particles. The Icing Cloud Environment (ICE) is made up of very small supercooleddroplets (10 to 50 micron diameters)which are sprayedinto the cold airstreamby specialnozzles (generallyhigh pressure,hot air and water). As the dropletstravel in the cold airstreamthey cool well below the freezingtemperaturewithout freezing (i.e., they supercool). Differentnozzlesor other devices are used to generate the larger droplets of freezingrain (FR) and the solid ice particles(SI) respectively. The cold airstreamis either cold ambient air or else it is cooled wholly (or in part) by a large refrigerationsystem. Other icing simulationtechniques.- Aerodynamicperformancepenalties caused by ice are traditionallyascertainedby flying the aircraftwith "plastic ice" shapes attachedto the wings and tail surfacesetc. This would be more difficaltto do safely with helicopterrotors. Analytical simulationsare also used to a large extent. Aircraft certificationis often based upon the similiarityof a new aircraftor componentto one

Original page 7

already certified. These methods are again not discussed in this paper; only icing simulationfacilitiesare discussed. Descriptionof Survey The surveywas limitedto those existingNorth American facilitiesthat have an icing simulationcapability. In other words, they producesuper- , cooled droplets (ICE and/or FR) in a cold moving air stream. Future icing simulationfacilitiesare also includeaif they are funded or seriouslyproposed. It is believedthat all of the facilitieswith ICE capabilityhave been included,but some low velocityfacilitieswith FR capabilitymay have been missed. . The facilitiessurveyed and their capabilitiesare listed in tables A to D, one table for each of the four types of facilitiesdescribedon figures 2(A) to (D). The capabilitiesof the individualfacilitieswere estimated by the technicalcontact person for that facility. Preliminarytables were completedby phone; later, the applicabletechnicalperson for each facility was sent a copy of the tables to check the entriesfor his facility. The numbers listed in the table are single point approximationsby him of the operatingcurves of that facility. Many of the capabilitieswere truncated so that comparisonswould not be made betweenfacilitieson the basis of unimportantexcess capabilitiesfor aircrafticing tests. Descriptionand Capabilitiesof Icing SimulationFacilities As noted before, the primarydifferencesbetween the varioustypes of facilitiesare in their geometry,airspeedand in the types of tests run. Each type of facility is now discussedalong with some comments about the capabilitiesof some individualfacilities. Wind tunnels.- The test section leg of a typical icingwind tunnel is schematicallysketched in figure 2(A). Table A lists the icingwind tunnels in North America. Most are closed loop wind tunnels;one is a Free Jet (entry A-5) that is listed in this table becauseit primarilydoes wind tunnel type of work. There are ten (10) tunnelsthat are active now; a very large wind tunnel has been proposed (A-lb),but facilities(A-4a) and (A-4b) have recentlybeen removed from the icingfacility rolls. The test sections of the existing tunnels range from 1.8x2.7meters for the largest(A-la) to 0.15 meters for the smallest (A-6a). The highestvelocityfor the larger existing tunnels is 470 km/hr; one small facility (A-5) can achieve M = 0.8. Most of the existing tunnels are limitedto sea level altitudes, except for the smallerones (A-5 and A-6b). All but (A-9) produce the Icing Cloud Environment(ICE) of adequatelysmall supercooleddroplets,including the 20 micron minimum goal. Only a few of the existing tunnels producethe larger droplets of freezingrain; none producesolid ice particles(SI). None produceSnow (S) either. It shouldbe pointedout that NRC researchers found that the best snow simulationwas made by "shovelingin" loosely packed natural snow. The LWC range, and the size of the uniform icing cloud are generallyadequate. All of these facilitieshave refrigerationso that they can be run all year; in addition,most are dedicatedto full time icing testing. The existingwind tunnels are ideallysuited for researchana developmenttype tests. Certificationtesting at the most severe icing conditionscan often be performed. But none of the wind tunnels can cover the entire FAR 25 envelope,or the entire altitudeand velocityrange of test aircraft. Furthermore,the tunnels are relativelysmall so that only

Original page 8

componentsof the aircraft are usually tested (e.g., inlets,tail section etc.). Helicopterrotors are simply too large. Icing scaling laws are often used to converttunnel results to the s!ze, airspeedand altitudeof the test aircraft(ref. 7), and to accountfor the deficienciesin the LWC and drop size of the cloud. Unfortunately,these scaling laws have not been adquatelyverified experimentally. Engine test facilities.- Table B indicatesthat there are 28 active engine test facilitiesthat can do engine icing tests; in addition,a large engine test facility (B-lc) is plannedfor 19B3. These are all engine test facilitiesthat do icing tests on a test engine as part of the test program; icing tests accpuntfor about 10 percentof the test programfor each engine. There are two basic types of engine test facilities: the Free Jet (fig. 2-B(a)) and the Direct Connect (fig. 2-B(b)). Many of the engine test facilitiescan be configuredto be run either way. In the Free Jet mode, the airstreamfrom the nozzle (i.e. the jet) passes around and through the engine. In the Direct Connectmode, the nozzle is extended to the engine inlet so that all of the airstreampasses throughthe engine. A number of these facilitiesare large and can attain high airspeedsand altitudes(e.g., B-la, B-lc, B-6b, and B-6c). The largestof the high speed Free Jets has a five foot diameter nozzle (B-l(b)). There is also a very large Free Jet (B-3) but it is limitedto very low velocities. The largestof the present engine facilitiesare too small to handle very largejet engines or large turboprops. Facilitiesthat can presently handle GIA propellerengines are limited in number (e.g., B-9). This problem is discussedin more detail in a later section. All facilitiesproduce an icingcloud environment(ICE). Only a few of the facilitiesproduce solid ice (SI) particlesnone producesnow. Most have refrigerationso that they can be run all year. Comparingall the capabilitiesof the Engine Test Facilitieslisted in table B with the Certificationrequirements,indicatesthat Certificationtests can be performedfor engines in most of these facilitiesover the entire FAR 25 certificationenvelope. The LWC in the cloud is reportedto be adequately uniform acrossmost of the flow. The Free Jet can be used for many icing experimentsthat would normally be performedin wind tunnels, especiallythose with test surfacesthat are short enough axiallyto stay within the potentialcore of the jet (cone shaped region of uniform velocity and low turbulencethat is about four nozzle diameterslong). The air speed and altitudecapabilityof some of these facilities(e.g. B-lb) are excellent. Low velocityfacilities.- There are 14 existingfacilitieslisted on table C. One will be mothballed by 1985 (C-1). All operate at a low velocity. All have FR capability;the first seven (7) can also produce the ICE. Most of thesefacilitiesare used for typicalcold room tests of equipment and personnelin a ground level environment(cold air at low velocity);aircraft icingtests are a small fraction of theirwork load. Most of the facilitiesare large refrigeratedcold rooms, where the test aircraftor component is tied down on the floor and subjectedto a fan blown spray (see fig. 2-C(a)). One of these refrigeratedfacilities(C-3a) is largeenough to permit a full scale aircraft to be tested with partial immersionin an icingcloud. Figure 2-C(b) describesthe unique HelicopterSpray Rig (C-1),which is located near Ottawa,Canada. In this case the test Helicopterhovers in the wind blown spray. A large engine test facility(C-2) has been listed here 6

Original page 9

in addition to being listed in table B. On Mt. Washington,equipment is tied down and subjectto severe natural icingconditions(C-5). The refrigeratedcold rooms can perform icin_tests all year, whereas facilities C-I, C-2, and C-5 are essentiallylimitedto winter operation. The LWC and drop size is adequate for ICE or FR tests (whicheveris applicablefor a given facility). Tankers for flight tests. - There are six (6) tankers listed in table D; one of these (D-5) has just been added, and (D-ib) is not yet in operation. Figure 2(D) describesthe HISS tanker (D-2) and its test helicopter;all other tanker facilitiesare fixed wing aircraft. There are differencesin the shape, size.and locationof the spraymanifold. Icing tests can be run with most fixed wing aircraft in any season by merely flying at the altitude where the desired temperatureoccurs. The limitedaltitude capabilityof helicoptersand some G/A aircraft limits the icing test season to the winter season. Most of the tankersare dedicatedto do icing tests full time. There have been many problemswith these facilities. One of the most serious was large droplets in the spray. Excessivelylarge droplets (larger than 100 microns) are usually easy to spot because the entire unheated nose of the aircraftwill ice up, whereas the small droplets in natural icing will only cause the small stagnationregion of the unheatedblunt nose to accumulateice. Tests were recentlyperformedon the spray nozzlesfrom the Army HISS and Air Force tankers in the NASA IRT (A-la). The presentmilitary tanker nozzleswere found to producedroplets that were 2 to 20 times too large, relative to the 20 micron goal. Fortunately,some of the nozzles tested producedthe desired droplet size at reasonableair and water pressures (ref. 8). Therefore,the droplet problemof the entire tanker fleet is on its way to a solution. The icing cloud from all of these tankers tends to be small and non-uniform,with the test aircraftweaving about within the icing cloud; this causes the LWC to vary with time. To partially accountfor this difficultya time averagedmeasurementof the LWC (e.g.,an ice accretionmeasurement)should be made at the locationwhere the critical ice accretionoccurs. Another problem is that most of these tankers are not readily available. The Flight systemstanker (D-5) is a recent welcomed addition to the fleet, because it is availablefor hire to all. Availabilityand Cost Availabilityand chargesvary greatly among facilities,and from test to test. The best recommendationis to first use the tables to select the facilitiesthat might fit your needs, then discussyour particulartest with the technicalcontact person (also noted in the tables) for each of those facilities. CURSORYEVALUATIONOF ICINGFACILITIES The adequacyof existing Icing SimulationFacilities,in performingthe varioustypes of icingtests listed in table 2, is judged in this section. Deficienciesare cited and some short term correctivemeasures are briefly discussed. These cursoryevaluationsare based upon: the data in the tables, additionalinformationand opinionsfrom the technicalpeople working with the facilities,and the partialevaluationsmade in references9 and 10.

Original page 10

Facilitiesto test instrumentation.- There are severalexcellent small governmentand company facilitiesfor R&D and certificationtests or icing instrumentation,(e.g.,A-4b, A-5, A-3, and A-7). The first two have the advantageof being able to cover a broad range of air speed, altitude and cloud conditions. In addition,the larger facilitiesoften can inexpensively run instrumentationtests along with anothertest. EnBine test facilities.- Table B indicatesthat there are many engine test facilitiesthat can do icing tests, and most of these have excellent capabilitiesfor testing engines over the whole FAR 25 envelope. Nevertheless there are some apparent deficiencies. Column 3 on table B indicates that there are xery few facilitiesthat can generateengine-aamaging-solid ice particles(from hail and snow to ice chunks). Snow, which is a problem for some inlets,can not be simulated in any facility. A facility is needed to test very large jet engines. This need should be satisfiedby the ASTF (see B-lc), which is planned to be built at AEDC in i983. The engine test facilitiesthat exist today are nearly all sizedfor turbofanand turbojet engines. A turbopropor G/A propellerenginewould be difficultto test in most of these facilitiesbecause of the prop size and the very large airflow that must be cooled in these once through-enginefacilities. Smaller turbopropscould be handled by some of the facilities(e.g., B-9, outdoor mode; B-3; and 3 meter diameter prop engines have been run in the diffuser of A-l). There is no facilityfor largehigh speed turboprops;howeverthe one proposed by NASA (A-lb) could again handle the task. Facilitiesfor fixed wind aircraft.- Certificationflight tests in natural icing are expensive but reasonablefor long range aircraftthat can fly to an area and altitudewhere icing is likely. But for short range aircraft (e.g.General Aviation),such flights are prohibitivelyexpensive. All fixed wing aircraftrequire simulationfacilitiesfor R&D icing testing and some certificationtesting. Short range aircraftuse simulation facilities,even for some of their certificationtesting. The best mix of icingfacilitiesfor the near term appear to be the fixed wing tankers (tableD) and the three ground facilitiesas outlinedbelow. RECOMMENDEDPRIMARYFACILITIESFOR FIXED WING AIRCRAFT (NEARTERM) Flight Tests t NaturalIcing Increased Long range aircraft: reasonablefor certificationtests cost Short range: only minimal programs affordable Flight tankers Ground Tests Full scale componentsat high speeds: (B-la) Full scale aircraft at very low speeds: (C-3a) Full scale componentsat moderate speeds: (A-l) and (A-2)

Original page 11

Tanker aircraftneed technicalimprovements(alreadydiscussed),additional experimentalverificationof the validity of this testingtechnique, and greateravailability. Three ground facilitiesare requiredfor R&D tests because no one existing facilitycovers the required range of size, air speedand altitude. For icing testingof full scale aircraft (but at very low velocities),a good choice is the Eglin cold rooms (C-3a). Full scale aircraftcomponents(e.g.,wings, inlets,etc.) can be tested at . moderate speeds in the NASA IRT (A-la) or the smaller tunnel at Lockheed (A-2). Most severe icing conditionsoccur at low speeds and low altitudes, where these facilitiesoperate. But if icing tests of full scale aircraft componentsare required at high speed and/or high altitude,then the AEDC free jet (B-lb).shouldbe considered. It should be pointed out that the large wind tunnel rehabilitationproposed by NASA (A-Ib) can handle all three requirementsof large size, high speed and high altitude;but this facility wouldn't be availableuntil 1987. Scaling laws are often used to compensatefor limitationsin the speed, altitudeor size of a facility,or limitationsin the icingcloud produced;however, the icing scalinglaws have not been adequatelyverifiedexperimentally. Facilitiesfor helicopters.- Performingflight tests on helicoptersin natural icing is extremelycostly,because the limitedrange and altitudeof the helicoptermakes it difficultto find icing conditions. Thereforeicing simulationfacilitiesare needed for the bulkof the icing tests; perhaps even includingcertificationtests (ref. 5). The engine, inlet and the fuselage componentscan be readilyhandled by existing engine test facilitiesand by the icing facilitiesused for fixed wing aircraft. The effect of the rotor can often be handledby using a reasonableangle of attack. This is not the case for the rotor. Icing test facilitiesfor the main rotor are not readily available,mainly becauseof its great size (12 to 18 meter diameter). Another difficultyis that the ice on the blades is subject to velocitiesranging from M = 0 to 0.8, and to largecentrifugal forces. A number of icing simulationfacilitiesand test rigs have been proposed and used to do rotor icing testing in the near term; these are listedand described in the followingtable in their approximateorder of: increasing experimentalcontrol and data confidence,and decreasingcost, but decreasing flight icing simulationaccuracy.

Original page 12

L TABLEOF ROTORICINGTESTMETHODSIN NORTHAMERICA(NEARTERM) Method Types of tests Problems A. Natural icing flights Helicopterflight tests Winter only. Hard to find _esired Ce.g., all companies) 0 icingconditions,so test programs expensiveand long before you get data of high confiaence. B. HISS Tanker (US army) Helicopterflight tests in spray Winter only. Somewhat less expencloud sive than naturalflights. It is assumed that the drop size and cloud size, etc. are made acceptable. Needs verification. C. Ottawa spray rig Helicopterflight tests at very Winter only. Closed after 1985. (NRC) low forward speeds in ground Relativelyinexpensive,relatively level spraycloud good control of conditions. Questions raised about simulationand using results for higher forward speeds.Needs verification. D. Large cold room Helicoptertie down tests at near All year. More expensivethan C. (Eglin AFB) zero forward speeds Many practicalproblems surfaced when it was tried. E. Scale model rotor in Ice accretionand its affecton Scaling laws for icinghave not icing wind tunnel rotor performance. Icing scal- been adequatelyverifiedexper- (NASA IRT) ing laws suggestthat this imentally.Smallestdrop sizes shouldwork. presentlyproduceablepermit models no smaller than 1/5 scale. Deicing systemscan not be scaled down readily. Needs verification. F. Rotor blade segment Main rotor blade segmmentof If you match G forces, the blade or tail rotor on a nearly full scale chord on velocitiesare very low. Needs rotating rig in icing rotatingrig to test deicing verification. Small tail rotors tunnel (NASA IRT) system operationand shedding should be closely simulated. at conservativeconditions. Also full to nearly full scale tail rotors G. Oscillatingor fixed Full scale blade segmenttested The maximum airspeedof the IRT is airfoil in icing tunnel with periodicor fixed angle of only M = 0.4, which is too low to (NASA IRT) attack (for ice accretion,aero- determineresults in the critical performanceand deicing system outer ha]f of the rotor. No G performance)in the absenceof G forces,thereforesheddingnot forces, but with blade bending true but conservative. Needs and twisting 10 verification.

Original page 13

Certificationtypes of tests tend to be performedon the facilities listed on the top half of the table, whereas R&D tests tend to be accomplished on the lowerhalf. But there is considerableuncertainty;verification tests are needed in order to determinewhere these simulationfacilities adequatelysimulatenatural icing on the rotor at the variousflight conditions. For example,the HISS tanker - until recently- generatedicing clouds with droplets that were about 10 times larger than those of natural icing (ICE); as a consequencethe icing resultswere closer to those produced by freezingrain (FR). The new spray nozzlesfor the HISS now produce the correct20 micron drop size (ref. 8). The Ottawa spray rig (B in the above table) is a valuablefacilitY;but it may not be availableafter 1985. It is also limitedto near-zerospeeds, althoughmany users have tried to extrapolatetheir results to cruising speeds with mixed success. The dynamic and aerodynamicdegradationof a rotor in icingcould be determinedin principleon a model rotor in an icingwind tunnel. Unfortunately icing tests of model rotors suffer from the followingdifficulties. First, existing tunnels are too small. Figure 3 indicatesthat the largest models that can be tested in the largestwind tunnel (A-la) with proper aerodynamicswould be i16 th to 1/12 th scale models. These models would require smallerdroplets,for proper scaling,than can made with present nozzles. Furthermore,these models would have to be built from scratchat very nigh cost, because the model rotors used by the helicoptercompanies are larger. The proposed largewind tunnel (A-ib) would be large enough to avoid this difficultybut this will not be availableuntil 1987. And finally, the icing scaling laws have not been validatedexperimentally. The test rigs for rotor icing are unable to simulateall the forces acting on the accreted ice. Specifically,a rotatingblade segment (F in the above table) will have very low blade velocitiesif the G forces are matched. Thus any test of a deicer systemwill be conservativeand will require analysis in order to relate the results to the actual conditionson a full scale rotor. The oscillatingblade rig (G in the above table) also gives a conservativesimulation,primarilybecause no G forces are Present. The main purpose, of the test rigs (F and G) are in the developmentof rotor icing and deicing analyses,and for conservativedevelopmenttests. The facilitiesand test rigs in the above table are, or will be, available soon. Two major facilitiesto do icing tests on full scale rotors have been seriouslyproposedfor the long term. One is an improvedtanker using either a large helicopteror a large slow speed fixed wing transport. The other is a very large slow speed test sectionfor facility (A-ib). The recent loss of the small high speedwind tunnel of NRC (A-4b) is a serious handicaptoward acquiringvital data on the aerodynamicdegradation caused by icing on 2D rotor airfoils. A replacementfor this facility is needed. 11

Original page 14

APPENDIX - ICING CLOUD MEASUREMENTS In the course of making this survey of facilities,a number of concerns about measurementswere brought out, which will be brieflydescribed in this section. They are: drople_ size measurements,the need for comparable liquid water contents (LWC) in all facilities,the uniformityof the LWC and drop size across the icing spray cloud, and the relativehumidity and temperature of the air within the cloud. LWC standard.- The survey broughtout the fact that it is difficultto compare icing resultsfrom differentfacilities. Part of the reason for this difficulty.isthat the accuracyof LWC instrumentsis often only about •20 percent. Another reason is that there is no standard instrumentthat can be used in all facilities. It has been suggestedthat a thin blade (0.32 cm thick x 1.9 cm chord) be used as an interim standardbecause it is adequatelyaccurateover a large range of conditions,and easy and inexpensive to make and use. The blade has been describedand investigatedin detail by Stallabrass(ref. 11). The blade is exposed to the cloud for only 30 seconds and the air is cold (<-12° C) in order to avoid thermal problems (ref. 11). The thicknessof the ice accretionon the thin edge is measured by a micrometer;the LWC is then determinedby a simplecalculation. With such a common standard,the LWC calibrationcurves for all facilitiesand LWC instrumentscould be inexpensivelymade consistent. A more accurate but more expensivestandardwould be a thin rotatingcylinder (0.32 cm diam rod) that is exposed to the cloud for a short time. Uniformityof LWC and drop size.- In all simulationfacilities,it is usually desirableto have a uniform LWC and drop size across the cloud. With reasonablecare in the design and maintainanceof spray nozzle arrays, the drop size should be reasonablyuniform,especiallywith small droplets. Achievinguniformityof the LWC is inherentlymore difficult. Becauseof turbulentmixing, the sprays from each nozzle undulate so that the accumulation of ice at a given point is due to the time varyingLWC from many nozzles. In tanker tests, the test aircraftalso undulateswithin the spray cloud. Forturnately,the ice build up is a time averagingprocess,which moderates this difficulty. Even in the well controlledwind tunnels, a few spray nozzlesmust be moved from time to time in order to keep the LWC across the cloudreasonably uniform. It proved to be difficultduring the surveyto quantitativelyestablishhow uniformeach facility'scloud was. Part of the reason for,this is that there is no standardfor uniformity. A practicaldefinitionof the uniform region would be wherever the time average LWC was within +/- 20 percent of the LWC at the center of the cloud. The uniformityof ground facility sprays is usuallydeterminedby the uniformity of the ice accretionon an array of cylinders. The determinationis based upon visual insPectionor a measurementof the uniformityof the mass of ice accreted. Droplet size. - A cursory look at column 12 in tables A through D indicates that the volume-mediandrop size is measured by a varietyof instruments, ranging from the "tried-and-true"older methods (e.g., rotating cylinders and oil slide) to modern methods (e.g., laser spectrometersand laser holographs). The relative accuracyand practicalityof these instruments is still being debated. The consensusof users (verbaland reports) indicatesthat the accuracyof the modern instrumentsappear to be +/- 3 to 6 microns, for the drop sizes typical of ICE (refs. 12 and 13). Most of the 12

Original page 15

old instrumentscan't give real time results, and they may not be quite as accurateas the modern instruments;on the other hand, they are far less expensiveto use if the amount of the data is modest. Two other points to consider are: How often must the drop size be measured,and what accuracy is required. For example, a 5 micron error might be acceptablein a Certificationtest requiring20 microns. In an icing simulationfacility, drop size measurementsneed not be made very often. Measurementshave indicated that the spray in the NASA IRT has producedthe same drop size for more than 20 years with only minimal maintenanceand demineralizedwater. " Relative humidity and temperature of the air in the spray cloud. - Accurate measurements of the relative humidity and temperature of Lhe air inside of the spray cloud are extremely difficult. The slightest amount of moisture will drive the relative humidity from 0 to 100 percent at the low air temperature of icing tests. Often, the best approach is to measure the conditions outside of the cloud and use a heat balance to calculate (ref. 13) the relative humidity of the air in between the droplets of the spray cloud. Recommendation. - A comprehensive experimental comparison of LWC, temperature, drop size and relative humidity instruments should be made in an icing tunnel, where the spray cloud is relatively repeatable. Such a test will determine the relative accuracy (not absolute accuracy) of the various instruments (both modern and old style) and their limitations for various applications. Along these lines it is also recommendedthat one small icing facility be used as the standard reference cloud, where instruments and their calibrations could me occasionally checked out. This would help standardize measurements made in ground facilities and in flight. This approach takes advantage of the repeatability of spray clouds in ground facilities, and admits that there may never be a cloud measurement whose accuracy is absolutely known. 13

Original page 16

REFERENCES 1. Pierre,M.; and Vaucherett,X.: Icing Test Facilitiesand Test Techniques. Aircraft Icing, AGARD Advisory Report 127, 1977,pp 6-2 to 6-5. 2. AirworthinessStandards,TransportCategoryAirplane. FederalAviation Regulationpart 25, Appendix C. 3. Werner, J. B., The Developmentof an Advanced Anti-lcing/Deicing Capabilityfor U.S. Army Helicopters,Vol. 1: Design Criteria and TechnologyConsiderations,USAAMRDL-TR-75-34A,1975. 4. Pfeifer, G. D.; and Maier, G. P.: EngineeringSummary of Powerplant IcingTechnical Data. PWA-5522,Pratt and WhitneyAircraft Group, 1977. (FAK-RD-77-76,AD-A045087.) 5. Minutes of Session V HelicopterIcing Syn_oosium,Ministry of Defense, London,Nov. 1978, pp. 335-339. 6. Newton, D.: A Review of the IcingSituationfrom the Standpointof GeneralAviation,Aircraft IcingNASA CP-2086,FAA-RD-78-109,1979, pp. 31-38. 7. Hauger, H. H.; and Englar,K. G.: Analysis of Model Testing in an Icing wind Tunnel. SM 14993, DouglasAircraft Co., 1954. 8. Belte, D. and Ferrel,K. R.: HelicopterIcing Spray System. American HelicopterSociety Paper 80-52,May, 1980. 9. Beheim,M.: Summary Report-lcingResearch and Facility Committee,Aircraft Icing. NASA CP-2086,FAA-RD-78-109,1979, pp. 121-i28. 10. Tests, Aircraft Ice Protection,FAA Advisory Circular,AC-20-73,1971, Chapter 4, pp. 19-27. 11. Stallabrass,J. R.: An Appraisalof the Single Rotating CylinderMethod of Liquid Water ContentMeasurement. Report LTR-LT-92,National Research CouncilCanada, 1978, Icing blademethod on Pages 10-18. 12. Jeck, R.: Performanceof the PMS AxiallyScatteringSpectrometer Probe, Aerosol Measurement,D. A. Lundgren,ed., Universityof Florida Press, 1979, pp. 294-311. 13. Hunt, J.: Engine Icing Measurement C_pabilitiesat AEDC. Icing Testing for Aircraft EnginesAGARD CP-236,1978, pp. 6-1 to 6-14. 14. Willbanks,C. F.; Schultz, R. J.: AnalyticalStudy of Icing Simulation for Turbine Engines in Altitude Test Cells. ARO-ETF-TR-73-59,Arnold Air Force Station, 1973. (AEDC-TR-73-144,AD770069.) 14

Original page 17

ABBREVIATIONS,AND FOOTNOTESFOR TABLES aTypes of icing and anti-deicingtests run: CPU - comp.letepropulsion unit; EDC - engine direct connect; FSC . full-scaleaircraftcomponent (includingwing, tail, fuselage,windshield,stores,gear, etc.); MS = model scale tests and instrumentation;IA = ice adhesion;CP = cloud physics;R = rotating experiments(e.g., helicopterrotor models and propellers);G = ground transportand installationsin freezing rain; FS = full-scaleaircraft;FLT = flight tests of aircraft;I = inlets with suction;P = complete propellerengines; H = human physiologicalexperi- - ments. bwhether simulated: ICE : icing cloud environment; SI : solid ice particles; FR = freezing rain; R = rain; Cparameter ranges vary with conditions; contact person. N = natural icing; S = snow. request operating envelopes from dModificationto do this has been seriouslyproposed. eTests in progress to extend these limits. 15

Original page 18

CAPABILITIES OF ICING SIMULATIONTEST FACILITIES IN NORTH AMERICA [CapabiLities estimated by technical contact person for each facility. ] A. WIND TUNNELS Fa- Facility name Types of Weather "l_pe of Size (see sketches), m Range of 'parameters used in icing tests c Instruments Technical Test Comment cility (Localioo) icing aimu- facility used for person to seano. tests run hted Test Uniform Air speed, Mla. total AHI- LWC_ Vol. med. local drop contact son chamber icing cloud km/hr air tern- rude, g/m drop size and perature (a) (b) °c A-1 NASA - Lewis Research Center (Cleveland, OH) m size, (LWC) m (a) IRT FSC, I ICE, Wind H = 1.8 hu = 0.9 I0 to 470 -30 0 e0.5 to ellto 25 Rot. cycls, andvar- J. Reinmann All Modernization MS, R, FH d tunnel W = 2.7 w u = 1.5 IA, pd L = 6 3.0 lonsmoderninstru- (216)433-4000 year nearly complete ments (rot. cyl.) (b) AWT- Rehabilitation FSC, I, IICE FR [ Wind D = 6 du = 4.5 I0 to M= 1.O -30 0 to 0.2 to 1O to 50+ Various modern J. Yuskn All Proposed for 1987 MS, R, !SI ' ' tunnel CPU, G, P D = 14 UP TO 95 15 000 3. (nozzles instruments (216)433-4000 year changed): A-2 Lockheed MS, FSC, ICE Wind H = 1.2 hu = 0.6 90 to 340 -20 0 0.7 to 10 to 25 Rot. eyls. B. Robinson All (Burbank, CA) tunnel W = 0.8 wu = 0.3 4.0 (rot. cyl.) (213)847-6121! year A-3 Hoeing lV, FSC, I ICE Wind H=0.5 hu=0.4 180to370 -30 0 0.3to 10toS0 Rot. cycls., oil R. Wilder All (Seattle, WA) tunnel W = 0.4 w u = 0.3 L = 0.9 (a) Large Tusnel 5.0 (nozzles slide (am. cyl.) (206)342-4776 year changed) / _ _ __ (b) High Speed MS, FSC ICE Wind H=W =0.3 hu =wu =O. 25 90 to M= 0.8 -30 0 to 0.2 to 15 to 25 Oil stide (rut. A. P'tce All To be mothballed tunnel 9 000 2. eyL) (813)993-2371 year A-5 AEDC Research Cell FSC, IV_ ICE Free Jet D = 0.9 du = 0.3 150 to -30 0 to 0.2 to 15 to 30 Various modern J. Hunt AH ......... --........... (Arnold AFS, TN) d = 0.3 et =0. 7 15 000 3. + instruments (615)455-2611 I year A-6 Rosemount (Minneapolis, MN) (a) Low Speed MS ICE Wind H = 0.15 hu ffiO.1 90 to 170 -30 0 0.2 to 20 to 40 Oil slide (rot. R. DeLeo All Rosemount use only tunnel W = 0.1 wu = 0.07 L=0.3 ! 1.5 cyl. ) (612)941-5560 lear (b) High Speed IV ICE Wind H = 0.15 hu = w u = 0.1 90 to 740 -25 0 to 0.1 to 1Oto 40 Oil slide (rot. R. DeLeo All Rosemount use only tunnel W = 0.3 3 0O0 3.0 cyl.) (612)941-5560 Tear L=0.8 A-7 FrostTunnel I, IA ICE Wind D = 9.5 du = 0.3 [0to240 -20 0 0.4to 20to50 Oilslide(rot. E.Gates All ..................... (Univ. of Alberts, home! (Octogano[ Canada) L = 0.9 3.0 (nozzles cyl.) (403)432-5180 Tear changed) A-8 ucLA Cloud Tunnel MS, CP ICE, R Vertical H = W hu ffiw u = 0.1 0 to 55 -39 0 0.1 to 2 to 50 Various modern H. Pruppacher All Free particle suspension wind = 0.15 tunnel L=0.5 3. instruments (213)825-1038 rear A-9 Army Natiek R&D G, FSC, H FR, R, S Wind H = 3 ........... 4 to 65 -30 0 10 em Not measured M. Kellberg All Mainly physiological (Natick, Mass.) tunnel W = 4.5 and Climatic Chamber L = 18 lower rain/ (rain gauge) (617)653-1000 year tests of humans hr .

Original page 19

B. ENGINE TEST FACILITIES • [Note that most free jets can do wind tunnel types of tests. ] Fa- Facility name Types of Weather Type of Size (see sketches), m Range of parameters used in icing tests c Instruments Technical Test Commen; cility (LocaUon) icing simu- facility used for person to seano. tests run lated Test Uniform Air speed, Min. total Alti- LWC. Vol. reed chamber icing cloud km/_r (a) (b) B-1 AEDC (Arnold AFS , "IN) (a) ETF EDC ICE Direct D = 3.7 Spray bars 0 to connect or 4.5 sized to M = 0.7+ d=1.5 L = 11 engine (b) Free Jet CPU, FSC ICE FreeJet D = 3.7 Spray bars 0 to I, MS d=l. 5 or 4.5 sized to M = 0.7+ L = 11 engine (c) ASTF CPU, FSC, ICE Free Jet D = 8 Spray bars 0 to I dffi2.7 L = 18 sized to M = 0.7+ engine B-2 Detroit Diesel Allison (Indianapolis, IN) (a) Comp. Test Inlet and ICE FreeJet D = 2.3 Spray bars 0 to M=0.7+ Facility compres- Direct L ffi9 sized to sor stage connect engine d= 0.5 (b) SneLl Engine EDC ICE Direct D = 0.45 Spray bars 0 to Facility connect [. = 1.2 sized to M = 0.7+ engine B-3 GE Cross-wind CPU, pal, ICE Free-Jet Outdoors du = 4.5 90 Facility R d outdoors (Peebles, OH) d= 7.0 B-4 P&W Altitude Facilities (E. Hartford, CT) (a) Large EDC, I ICE Direct D=5.5 Spraybars 0toM=0.5 connect L = I0 sized to engine (b) Smaller EDC, I ICE Direct D = 3.7 Spray bars 0 to M=0.5 connect sized to engine (c) P&W Sea Level EDC ICE Direct Varies Spray bars 0 to Facility connect with test sized to M = 0.5 celis engine local drop contact son air tern- hide, g/m 3 drop size and perature, m size, (LWC) °C um -30 0 to 0.2to 15 to30 Various modern J. Hunt ALl ...................... 15 000 3.+ instruments (615)455-2611) year -30and 0 to 0.2to 15to30 Various modern J. Hunt All ...................... lower 15 00C 3.+ instruments (615)455-2611 year -30 and 0 to 0.2 to 15 to 30 Various modern W. Bates All Planned for 1983 lower 15 000 3.+ instruments (615)455-2611 year -30and 0 0. 2to 15to40 Rotating W. Stiefel All ...................... lower 0 to 3.5 cylinders (317)243-4066 year 6 000 -30and 0 to 0.2 to 15to40 Rotating W. Stiefel All ...................... lower 6 000 3.5 cylinders ($17)243-4066 year Ambient 0 0.4to 15to50 Knoltenberg R. Keller Win....................... 'air to 3.5 spectrometer (513)243-4483 ter -20 (rot. cyl.) -25 0to 0.2to 15in40 Ollslide J. Barlock All ...................... 6 700 9.0 (203)565-2091 year -30and 0to 0.2to 15to40 Oil slide J. Barloek All ...................... lower 6700 9.0 (203)565-2091 year -20 0 0.2 to 15 to 40 Dil slide J. Hartock Win....................... (am- 9.0 (203)565-2091 ter blent) B-5 McKinley Climatic Lab CPU, FSC ICE, SI Fan blown H = 7.5 hu =3 0 to (30 to '1o) -3b and 0 0.1to 12to60 Particle inter- R. Toliver All ...................... Engine Test Cell -- spray W = 9 w u = 6 (Eglin AFB, FL) FR, R indoors L = 40 lower 3. _ ferometer (904)882-3626 rear 800to 150C nozzles (rut. cyl. ) changed)

Original page 20

B. Concluded. ENGINE TEST FACILITIES [Note that most free jets can do wind tunnel types of tests. ] Fa- Facility name 3_tpes of Weather Type of Size (see sketches), m Range of parameters used in icing tests c Instruments Technical Test Comment :illty (Location) icing simu- facility used for person to seano. testa run lated Test Uniform Air speed, Min. total ALii- LWC. Vol. reed localdrop contact son chamber icing cloud km/hr air tem- rude, g/m 3 drop size and (a) (b) B-6 Naval Air Propulsion Facility (Trenton, NJ) (a) Five small engine EDC, CPU, ICE, SI, Free Jet H = W = : Spray bars 0 in cells I, FSC, 1V_FR, R d = 0.6 L = 6 sized to M = 0.7+ engine CO)Two large sea EDC, CPU, ICE, SI, Free jet H = 4.5 Spray bars 0 to level cells I, FSC, .MS FR, R d = 1.2 W = 7 sized to M = 0.7+ L = 17 engine (c) Three large EDC, CPU ICE, SI Free Jet D = 5 Spray bars 0 to altitude cells I, FSC, !V_ FR, R d = 1.2 L = 9 sized to M = 0.7+ engine B-7 Teledyne Altitude Cells (Toledo, OH) (a) Chamber 1 CFU, EDC ICE, SI Free jet D = 2. 7 Spray bars 0 to FR, R or direct L = 5 sized to M = 0.7+ connect, engine (b) Chamber2 CPU, EDC ICE, SI, d=0.2 H=2.5 Spraybars 0to FR, R W = 2.5 sized to M = 0.7+ L = 4 engine B-8 Avco Lyeoming (Stratford, CT) (a) Component EDC ICE, FR Direct Spray bars 0 to 370 Facility connect sized to d = 0.4 engine Co)Engine Test EDC ICE, FR d = 0.4 W = 3.7 Spray bars 0to200 Facility H = 2.7 sized to d = 1.2 Outdoors engine B-9 NRC, Ce11#4 EDC, CPU ICE,SI Free Jet HfW Spraybaro 0to650 (Ottawa, Canada) or direct = 7.5 sized to connect engine d= 0.75 CPU, P d = 2.0 0 to 93 B-10 GarretlcingFacliLiies: CPU, I ICE, ISI Free Jet H=3 Spraybars M= (Phoenix, AZ) EDC, or direct W = 4 sized to 0.01 - 0.7 Cell 1, Cell 2, and FSC, P connect, L = 10 engine Cell 3 d = I. 0 to •1 perature m size, (LWC) °C um -30and 0in 0. I to 15 to50 Knollenberg Resource All ...................... lower 15 000 2. (nozzles spectrometer Mgr. year changed) and OAF (609)896-5655 (rot. cyl.) -30and 0 0. lto 15to50 Knollenberg Resource All ...................... lower 2. (nozzles spectrometer Mgr. year changed) and OAP (609)896-5655 (rot. cyl. ) -30and 0to 0.1to 15to50 Knollenberg Resource All ...................... lower 15000 2. (nozzles spectrometer Mgr. _ear changed) and OAP (609)896-5655 (rot. cyl. ) - 30 and 0 to Up 15 to 25 Oil slide R. Trauth All lower 15 000 to 3. (419)470-3236 rear -30and 0to Up 15to25 Rotating R. Trauth All lower 15 000 to 3. cylinders (419)470-3236 rear - 30 and 0 0.1 to 15to 40 Oil slide J. Sherman All lower 3. (rot. cyl.) (203)378-8215 year -30and 0 0. lto 15to40 Oil slide J. Sherman All lower 3. (rot. cyl.) (203)378-8215 year - 20 Winter -20and 0 0.2to 15to40 Ollslide W. Grabe Win....................... lower 2. (rot. cyl.) (613)993-2214 ter Am. blent -30and 0to 0.1t¢ 10toS0 Rot. cyls. J. Pyne All ...................... lower 15 000 6.0 (Rot. cyls.) (622)267-3853 rear

Original page 21

C. LOW VELOCITY FACILITIES Fa- Facility name _rpe of Weather ype of Size (see sketches),, m Range of parameters used in icing tests c Instruments Technical Test Comment cllity (Location) icing simu- facility used for person to seano. tests run laied Test Uniform Air speed, Min. total Alti- LWC, Vol. reed. local drop contact son chamber icing cloud km/hr (a) (h) air tern- tude, i g/rn3 drop size and perature rn size, (LWC) °C m C-1 NHC Helicopter FLT(hel- ICE, FR Windblown D= o Spray Ambieniwind, -20 0 '0.1to 30to60 Oil slide T. Ringer Win- To be mothballed Spray Rig icopters spray manifold 20 to 45 (Ottawa, Canada) in hover) outdoors hs = 4.5 (gusty) ws = 23 C-2 G.E. Cross Wind CPU, P d, ICE, FR Free Jet D= du=4.5 90 Facility Rd outdoors (Peebles, OH) C-3 McKinley CHma_c Lab (Egltn AFB, FL) (a) Main Chamber FS, Rd ICE, SI Fan blown H = 21 Spray 0 in (30 FH, H spray W = 78 manifold to 75e) indoors L = 76 hs = 3 (depending w s = 9 on LB) (b) Engine Test Cell CPU, FSC ICE, SI Fan blown H = 7.5 Manlfold 0 to (30 FR, R spray W = 9 h s = 3 to 75) indoors L = 40 ws = 6 (depending on LB) (c) All Weather Room FSC ICE, SI Fan blown H = 4.5 Manifold 0 to (30 FR, R spray W = 0.5 h s = 3 to 75) indoors L = 12 ws = 3 (depending ouLB) C-4 U.S.Army CRREL FSC, M ICE,SI, Fan blown H = 1.1 I0 to 20 ColdRoom H, IA FR, R spray W = 0.7 (Hanover,NH) indoors L = 1.5 C-5 Mt. Washington FS, CP, Natural icing of tied down equipment on top 0 in 180 Observatory MS of mountain (gusty) (Gorham, NH) C-6 U.S. Navy PMTC . FSC, R, FR Fan blown H = 7.6 hs =w s 0 to 75 (Pt. Magu) FS, G spray W = L = 1.2 Climatic Hanger indoors = 18 S • C-7 ActonEnvironmental G R FR, Fan blown H = 6 ds = 2.5 0 to45 TestCorp. S spray W = 4.5 (Acton,Mass.) indoors L = 7.5 C-8 NRC G, FS FR, Sd Fan blown H= 4.3 ds = 1.2 0 to 55 (Ottawa, Canada) spray W = 4.5 to 2.5 Cold Chamber #1 indoors L = 15.2 NHC G, FS FR Fanblown H = 5 ds = 1.8 0 in 55 (Ottawa, Canada) spray W = 5 Cold Chamber #2) indoors L = 7 C-9 Wyle Labs G, FSC FR Fan blown H = 5 0 to 35 (Norco, CA) : spray W = 4.5 Cold Room indoors L = 11 C-10 Arcte'c C_mnda Lid. G, IA FHd, S d Fan blown H = 3.7 to 35 (Ottawa, Canada) spray W = 5.5 Cold Room indoors L = 9 (am- 0.8 (rot. cyl.) (613)993-2439 ter in 1985 blent) -20 0 0.4to 15in50 Knnlienberg R. Keller Win- (am- 3.6 spectrometer (5137243-4483 ter blent) (rot. cyl.) - 30 and 0 0. I to 12 to 60 Particle inter- R. Toliver All Largest cold room lower 3 800to ferometer (904)882-3626 year 1500 (nozzles (rot. cyl.) changed) -30 and 0 0.1to 12 to 60 Particle inter- H. Toliver All lower 3 8015000to ferorneter (9047882-3026 year (nozzles (rot. cyl.) changed) - 30 and 0 0. Ito 12 to 60 Particle inter- H. Toliver All lower 3 8150000to ferometer (904)882-3626 year (nozzles i (rot. cyl. ) changed) -30 and 0 1to 10to60 Cascade G. Ashton All lower 2.5 impactor (603)043-3200year - 20 and 1800 GeneraUy severe Rotating J. Howe Fall lower natural conditions cylinders (603)466-3388 to spring -30and "0 30 cm 500 to Oil slide D. Everett All Inwer rain/ 4500 (rain gauge) (805)982-8011 year hr 5 cm 50to 100 snow/ hr -30and 0 10 crn 1000to Notmeasured H. Gilfoy All lower rain/ 4000 (raingauge) (617)263-2933year hr -30 and 0 0.3 crr 500 to Screen method T. Ringer All lower rain/ 1000 (accumulation (613)993-2439 year hr rate) -30 and 0 0.3dcrn ! 500to Screen method T. Ringer All lower rain/ 1000 (accumulation (613)993-2439 year hr rate) -30 and 0 :12 cm ...................... M. Clark All ..................... lower rain/ (714)737-0871 year 'hr -30 and 0 ...... A. Nawwar All ..................... lower (613)592-2830 year

Original page 22

D. TANKERS FOR FLIGHT TESTS [In addition, most airframe companies can test aircraft in natural icing. ] Fa- Facility name _pes of Weather Time in Size of spray, m Range Of parameters used in icing tests c Instruments Technical Test Comment ciLity (Location) icing simu- icing at no. tests ran lated high At nominal Manifold Air speed, LWC, distance km/hr rain LB IAS (a) (b) D-I Air Force (Edwards AFB, CA) (a) KC 135 Tanker Fit. ICE, N 80 At ds = 1.2 300 to 650 LB = 60 (370 nora.) R, FR d=3 (b) C 130 Tanker Flt. ICE, N 60 At ds = 1.2 190 to 390 LB = 60 (280 nora. ) R, FR d= 5 D-2 ArmyHISS HeLicopter Fit. ICE, N 30 At ihs = 1.6 110to140 used for person to season Mtn. total Altl- LWC, VoL reed. local drop contact (find air tem- rude g/m 3 drop size and temp. perature m size, (LWC) at °C /_m altitude) -20 1200 0.051o 28to35 Knollenberg R. Morrison All year Final calibration (ambient) in 1.5 spectrometer (805)277-3068 in 1931 8000 0.5to 200 to 800 ( " ) 32; -20 1200 0.05to 28to35 Knolienberg R. Morrlson Aliyear Planned for 1981 (ambient) to I. 5 desired spectrometer (805)277-3068 8000 0.05to 200tn800 ( ,w ) 32. -20 600 0. lto 25to30 KnnUenberg C. Franken- NormaLly Testing to increase Tanker LB = 50 ws = 12 (120 nom. ) (ambient) to 1.0 desired spectrometer berger winter cloud size (Edwards AFB, CA) h = 3 w= 12 D-3 C_ssna 404 Tanker Fit. ICE, R, 60 At ds = 0.6 165to330 (Wichita, KA) FR, N LB = 150 (V-bar) (260 nora.) d = 6 D-4 Piper Cheyenne Tanker Fit. ICE, FR, 14 At hs = 1.2 200to300 (Lock Haven, PA) R,N LB=30 Ws= 1.8 (240nora.) h = 3 w=5 D-5 Flight Systems T-33 Fit. ICE, R, 45 At hs = 0.3 230to420 Tanker FR, N LB = 60 ws = 0.9 (370 nora.) (MoJave, CA) d = 2.5 3500 (Letgh) (805)277-2271 -20 300 0.05to 20to49 Gelatin sLide D. Hazelwood All year (ambient) to 4.0 (water (J&W) 1316)946-6606 8000 nozzles) -20 300 0. lto 30to50 Gelatin slide J. Bryerton Not (ambient) to 1.7 (J&W) '.717)748-6711 summer 8000 -20 300 0.1 to 17to50 Knollenberg J. Ltgon All year (ambient) to 1.0 spectrometel (805)824-4801 8000 ( " )

Original page 23

O \ ,-STRATIFORM CLOUD -10-- \ ,-,- CUMULIFORM -20-- -30 I l I I 2 4 6 8 ALTITUDEkm, 3 2 TEMPERATURE, °C "_ INTERMII-rI_ITMAX. G (CUMULIFORMCLOUDS) _ \ o \ i 0 20 30 40 DROPSIZE,microns Figure1. - FAR.- 25icingcertificationconditions;99.9_ exceedanceprobability,ref. 3.

Original page 24

SCHEMATICSKETCHESFROMSIDEVII_N /-SPRAY MANIFOLD /, r_-__ i r_° _ ,, OFAiCCOMPONENT /- TESTMODEL -.'L:,;: l_i--",;.__ "'"- - -_ .... "...... "---"----TEST SECTION,U ,,- A. WINDTUNNELS /--SPRAY I MANIFOLD D r-TESTENGINE I OR MODEL / _ / I_>';i_v"':,-'., r,, ]:- - .-:l I. J,';',I',;;'I-,,..-,1-...,,,;.-.,,.,1:,_.;./-. \ x...NOZZLEt (a)FREEJET. (b)DIRECTCONNECT. B. ENGINTESTEFACIUTIES Figure2. - Typesoficingsimulationfacilities.

Original page 25

te \ -SPRAY MANIFOLD \ \ .-ARRAY OFFANS {a)FANBLOWNSPRAYINA LARGEROOMOROUTDOORS. -- VWlND SPRAYMANIFOLD- h s (b)WINDBLOWN SPRAYOUTDOORS. L. H LI_, _ s, ,, C. LOWVELOCITYFACIUTIES LB TESTAIC"_ D. FLIGHTTESTSWITHTANKER Figure2. - Concluded.

Original page 26

u., 20 18-- 2Op_IATURAL / ,._ 16--- CLOUDS) / - PRESENTCLOUD _ 12_ SIMU ,< lO ................. _ 8 J!° - / ToS,MU,ATDROPSEZ0. _ 6J OFNATURALCLOUDS m 4-- } I i I ii " 11121/10 118 116 il14 1 I I I i CH4/UHIH B01_ I J J t [ CH47UHIH B0105 MODEL/FULLSCALEFACTOR i i 112 1/1 MODELROTORI,5 M--ISIZE:RT 3.SM--AWT Figure3. - Effecton dropsizeon scaling(oldstyleinstruments).

Original page 27

Original page 28

I, Report No. 2. GovernmentAcc_sion No. 3. Recipient's CatalogNo, NASA TM-81707 4. Title end Subtitle 5. Regort Date SURVEY OF AIRCRAFT ICING SIMULATION TEST February 1981 FACILITIES IN NORTH AMERICA 7. Author(s) • • William Oisen 9. PerformingOrganizationName endAddress National Aeronautics and Space Administration Lewis Research Center Cleveland, Ohio 44135 12. Sl_nsoring Agency Name and Address NationalAeronauticsand Space Administration Washington, D.C. 20546 15. Supplementary Notes 16. Abstract S. Performing505-44-12OrganizationCode 8. PerformingOrglnitetion Report No. E-736 10. Work Unit No, 11. Contract or Grant No. 13. Type of Report and PeriodCovered Technical Memorandum 14.SponsorinAgegncyCode A surveywas made of theaircrafticingsimulationfacilitiesin North America. This was requestedof NASA by several committees concernedwith AircraftIcing. A similarsurvey of European facilitiehsad alreadybeen reportedin AGARD advisoryreport127. There are 12 wind tunnels,28 enginetestfacilities6 a,ircrafttankersand 14 low velocityfacilities,thatcan perform variousaircrafticingtestsfullor parttime. The survey determinedthelocationand sizeof the facility,itsspeed and temperature range, icingcloudparameters, and the technical person to contact. These resultsare presentedin tabularform. The capabilitieosf each facilitwyere estimatedby itstechnicalcontactperson. The adequacy ofthese facilitiefsorvarious typesof icingtestsis discussed. 17. Key Words (Suggestedby Author(s)) 18. Distribution Statement Test facilities Aircrafticing Unclassified- unlimited STAR Category 03 19. Security Classif.{of this report) 20. SecurityClassif,(of this page) 21. No. of Pages 22. Price" Unclassified Unclassified * ForsalebytheNationalTechnicalInformationService,Springfield.Virginia22161

Original page 29

Original page 30

National Aeronautics and SPECIAL FOURTH CLASS MAIL Postage and Fees Paid Space Administration BOOK. National Aeronautics and Washington, D.C. 20546 Official Business Penalty for Private Use, $300 I_J/I_A Space Administration NASA451 • - J PosThAsTE.,Postal,_°Manua..l),,Do***NotReturnab,.,s. Ib