Philip CE Jorgenson Joseph P Veres Shashwath R Bommireddy and S - PDF document

1 Philip C.E. Jorgenson, Joseph P. Veres,
Philip C.E. Jorgenson, Joseph P. Veres, Shashwath R. Bommireddy, and Samaun NiliGlenn Research Center, Cleveland, Ohio ed ResearchEngine (HURE) With Ice Crystal Cloud Ingestionat Simulated Altitudes: Public Version NASA STI Program . . . in Pro leSince its founding, NASA has been dedicated The NASA Scienti c and Technical Information (STI) Program plays a key part in helping NASA maintain The NASA STI Program operates under the auspices of the Agency Chief Information Of cer. It collects, organizes, provides for archiving, and disseminates NASA’s STI. The NASA STI Program provides access to the NASA Technical Report Server—Registered (NTRS Reg) and NASA Technical Report Server—Public (NTRS) thus providing one of the largest the world. Results are published in both non-NASA channels and by NASA in the NASA STI Report • TECHNICAL PUBLICATION. Reports of cant phase of research that present the results of NASA cant c and technical data and information NASA counter-part of peer-reviewed formal • TECHNICAL MEMORANDUM. Scienti c ndings that are preliminary or of • CONTRACTOR REPORT. Scienti c and ndings by NASA-sponsored • CONFERENCE PUBLICATION. Collected c and technical • SPECIAL PUBLICATION. Scienti c, NASA programs, projects, and missions, often • TECHNICAL TRANSLATION. English- c and technical material pertinent to NASA’s mission.For more information about the NASA STI • Access the NASA STI program home page at http://www.sti.nasa.gov• E-mail your question to help@sti.nasa.gov• Fax your question to the NASA STI • Telephone the NASA STI Information Desk at 757-864-9658• Write to:NASA STI Program Mail Stop 148 NASA

2 Langley Research Center Hampton, VA 2368
Langley Research Center Hampton, VA 23681-2199 Philip C.E. Jorgenson, Joseph P. Veres, Shashwath R. Bommireddy, and Samaun NiliGlenn Research Center, Cleveland, Ohio ed ResearchEngine (HURE) With Ice Crystal Cloud Ingestionat Simulated Altitudes: Public VersionNational Aeronautics and This work is supported by the Advanced Air Transport Technology Project in the NASA Advanced Air Vehicles Program, and is in response to the Engine Icing Technical Challenge. We would like to acknowledge Ashlie Flegel for leading the experimental research test and providing data for the analysis, Dr. Judith Van Zante for providing icing cloud expertise, Dr. William B. Wri(Vantage Partners, LLC.) for guidance in providing the MELT code subroutine, and Dr. Jen-Ching Tsao (Ohio Aerospace Institute) for his insights. The authors would also like to thank the entire NASA PSL staff for their dedication and support of this test. The Available fromTrade names and trademarks are used in this report for identi cation only. Their usage does not constitute an of cial endorsement, either expressed or implied, by the National Aeronautics and : This material has been technically reviewed by technical management. NASA STI ProgramNASA Langley Research CenterHampton, VA 23681-2199National Technical Information Service eld, VA 22161This report is available in electronic form at http://www.sti.nasa.gov/ and http://ntrs.nasa.gov/ This work was sponsored by the Advanced Air Vehicle Program at the NASA Glenn Research Center ��NASA/TM220023ContentsAbstractNomenclatureIntroductionCompressor Code CalibrationFanStator Analysis; Risk of Accretion Gover

3 ned by the Icing Wedge ThresholdsSplitte
ned by the Icing Wedge ThresholdsSplitterLipStrut Figure 12.Ice accretion occurred in the fanstator during Escort 121. The static wet bulb temperature was within the Icing Wedge.Figure 13.Computational Process for NonAdiabatic Walls that was incorporated into the COMDESMELT code to estimate the core wall temperature at the splittershroud region.Figure 14. ��NASA/TM220023Figure 16.Calculated wall temperature versus IWAR of the HURE splittershroud region for data points with ice accretion at high altitudes and at 5K ft, compared to the measured wall temperature of the LF11 engine (Ref. 15).Figure 17.Escort 153. Ice accretion with ice horns on the splitter.Figure 18.Ice accretion occurred at the splitter shroud during Escort 153. The static wet bulb temperature is 19 °R below the Icing Wedge minimum threshold of 492 °R.Figure 19.For Escort 156, there was ice accretion on the shroud, without horns.Figure 20.Ice accretion occurred at Escort 156. The static wet bulb temperature is 24 °R below the Icing Wedge minimum threshold of 492 °R.Figure 21.Calculated wall temperature versus IWAR of the HURE splittershroud region with no ice accretion, only ice collection and rapid shedding. Black symbols are the measured wall temperaturesat the location of ice accretion (Ref.15).Figure 22.Ice collection on the splitterlip with rapid shedding, is not considered ice accretion since it does not grow (Escort 276).Figure 23.Ice accretion did not occur for Escort point 276 due to additional heat from the warm wall at the statorshroud region. The static wet bulb temperature there was on the order of 487.9 °R.Figure 24.The s

4 tatic wet bulb temperature (Twbs) versus
tatic wet bulb temperature (Twbs) versus IWAR in the splittershroud region, for the cases with hard ice accretion and ones which had only collection, but no ice accretion (from able 4 and Table 5).Figure 25.Ice did not collect on the splitterlip or the shroud, at data point. (Escort 289).Figure 26.Ice accretion did not occur in Escort 289 which had a static wet bulb temperature 31 °R below the Icing Wedge minimum threshold of 492 °R.Figure 27.Calculated wall temperature and the static wet bulb temperature at the splittershroud wall region at 5K ft altitude for ice accretion and no ice accretion data points (from Table 8 and Table 9).Figure 28.Splittershroud: 5K altitude ice accretion on splitterlip and shroud for Escort data point 243.Figure 29.The static wet bulb temperature distribution and particle melt ratio for the 5K ft altitude Escort point 243 through the full fan, the fancore and the four stage axial compressor, as well as the particle melt ratio.Figure 30.Calculated wall temperature at the splittershroud wall region versus calculated air static temperature.Figure 31.Measured IGV metal temperatures versus IWAR for all operating points with video confirmation.Figure 32.Ice accretion on the IGV surface near the tip (Escort 284 at 36K ft altitude).Figure 33.The static wet bulb temperature distribution and particle melt ratio for the 36K ft altitude Escort point 284 through the fancore and the four stage axial compressor, as well as the particle melt ratio.Figure 34.Ice collection on the IGV leading edge near the tip withrapid shedding (Escort 123 at 40 K ft altitude).Figure 35.The static wet bulb temperature di

5 stribution and particle melt ratio for t
stribution and particle melt ratio for the 40K ft altitude Escort point 123.Figure 36.Computed ice particle temperature and measured IGV metal temperatures versus IWAR. The black and blue symbols represent ice accretion or collection. The red symbols represent points without ice accretion.Figure 37.Measured IGV metal temperatures pre and post ice cloud on versus IWAR. The black and blue symbols represent ice accretion or collection. The red symbols represent data points without ice accretion. ��NASA/TM220023Figure 38.Computed static air temperature and measured IGV metal temperatures versus IWAR. The blackand blue symbols represent ice accretion or collection. The red symbols represent data points without ice accretion.Figure 39Computed static air temperature and ice particle temperatures versus IWAR after ice cloud on. The black and blue symbols represent ice accretion or collection. The red symbols are data points without ice accretion.Figure 40.Comparison of the IGV measured metal temperature to the calculated metal temperature at the splittershroud region. The trend of the calculated metal temperature closely tracks the measured IGV metal temperature.Figure 41.Percentage difference between the CD model and theEscort test data as a function of Escort number.List TablesTable 1.FanStator: No Ice AccretionTable 2.FanStator: Ice AccretionTable 3.Data Points Utilized to Calibrate Heat Transfer Model in COMDESMELTTable 4.SplitterShroud: Ice Accretion Data PointsTable 5.SplitterShroud: Ice Collection with Rapid Shedding; No Accretion (Calculated Hot Wall)Table 6.SplitterShroud: No Ice Accretion, Except at Sup

6 port Strut (Low Twbs)Table 7.SplitterShr
port Strut (Low Twbs)Table 7.SplitterShroud: No Ice Accretion (Calculated Hot Wall)Table 8.SplitterShroud: 5K Altitude; Ice AccretionTable 9.SplitterShroud: 5K Altitude; No Ice AccretionTable 10.Inlet Guide Vane: Ice Accretion; Video ConfirmationTable 11.Inlet Guide Vane: Ice Collection with Rapid Shedding; Video ConfirmationTable 12.Inlet Guide Vane: No Ice CollectionNo Ice Accretion; Video ConfirmationTable 13.let Guide Vane: No Video Available for ConfirmationTable 14.HPC Stage 1; Stator: No Video Available for Confirmation ��NASA/TM220023Analysis of the Honeywell Uncertified Research Engine (HURE) ith Ice Crystal Cloud Ingestion at Simulated AltitudesPublic VersionPhilip C.E. Jorgenson,Joseph P. Veres, Shashwath R. Bommireddy,and Samaun NiliNational Aeronautics and Space AdministrationGlenn Research CenterCleveland, Ohio 44135AbstractThe Honeywell Uncertified Research Engine (HURE), a research version of a turbofan engine that never entered production, was tested in the NASA Propulsion System Laboratory (PSL), an altitude test facility at the NASA Glenn Research Center. The PSL is a facility that is equipped with water spray bars capable of producing an ice cloud consisting of ice particles, having a controlled particle diameter and concentration in the air flow. In preparation for testing of the HURE, numerical analysis of flow and ice particle thermodynamics was performed on the compression system of the turbofan engine to predict operating conditions that could potentially result in a risk of ice accretion due to ice crystal ingestion. The results of those analyses formed the basis of the test matr

7 ix. The goal of the test matrix was to h
ix. The goal of the test matrix was to have ice accrete in two regions of the compression system: region one, which consists of the fanstator through the inlet guide vane (IGV), and region two which is the first stator within the high pressure compressor. The predictive analyses were performed with the mean line compressor flow modeling code (COMDESMELT) which includes an ice particle model. Together these comprise a onedimensional icing tool. The HURE engine was tested in PSL with the ice cloud over the range of operating conditions of altitude, ambient temperature, simulated flight Mach number, and fan speed with guidance from the analytical predictions. The engine was fitted with video cameras at strategic locations within the engine compression system flow path where ice was predicted to accrete, in order to visually confirm ice accretion when it occurred. In addition, traditional compressor instrumentation such as total pressure and temperature probes, static pressure taps, and metal temperature thermocouples were installed in targeted areas where the risk of ice accretion was expected. The current research focuses on the analysis of the data that was obtained after testing the HURE engine in PSL with ice crystal ingestion.The computational method was enhanced by computing key parameters through the fanstator at multiple spanwise locations, in order to increase the fidelity with the current meanline method. In addition, other sourcesof heat (nonadiabatic walls) were suspected to be the cause of accretion near the splitterlip and shroud. A simple order of magnitude heat transfer model was implemented into the COMDE

8 SMELT code to estimate the wall temperat
SMELT code to estimate the wall temperature minimum and maximum thresholds that support ice accretion, as observed by video confirmation. The results from this model spanned the range of wall temperatures measured on a previous engine that experienced ice accretion at certain operating conditions.Future analyses will require a higher fidelity thermal analysis of the compression system metal walls to accurately calculate the total heat flux to the ice particle. For many data points analyzed, there were differences between the thermodynamic system model and the measured test data that may partially be responsible for uncertainties with the results of the current analyses. Summer Intern in Lewis’ Educational and Research Collaborative Internship Project (LeRCIP) ��NASA/TM220023NomenclatureAlt altitudecustomer deckCFMcubic feet per minuteCOMDEScompressor flow analysis codeDeltaISAoffset temperature from ISAHPChigh pressure compressorIGVvariable inlet guide vane ISAInternational Standard AtmosphereIWARicewater flow rate to air flow rate ratioIWCice water content (glaciated ice crystals), g/mleading edgemass flow rate of air, lbm/sMELTice particle thermodynamic state modelN1, N2fan, HPC rotational speedsPT1engine inlet total pressureair volumetric flow rate, CFMTambambient temperature, statictrailing edgeTT1engine inlet total temperatureTwbsstatic wet bulb temperatureTWCtotal water content, g/mmicronIntroductionIce crystals ingested into turbofan engines during the operation of commercial aircraft at high altitudes in high ice water content environments can result

9 in ice accretion in the compression sys
in ice accretion in the compression system (Refs. 1 and2). As ice crystals are ingested into the fana portion of the ice crystalsmelt due to the rising static temperature of the air. It is hypothesized that this icewater mixture then impacts and cools the surfaces through evaporation and ultimately ice accretes on the components. The accreted ice can cause one or more of the following modes of failure: uncommanded loss of thrust control, compressor surge or stall, ice shedding which can result in structuraldamage to the compressor blades, and possible combustor flameout. To improve understanding of the causes of ice accretion within an engine, full engine testing with ice crystal ingestion has been performed in the NASA Propulsion Systems Laboratory (PSL) (Refs. 3to 5), as well as fundamental ice crystal testing (Refs. 6 and7).This study focuses on the analysis of the test data obtained from the Honeywell Uncertified Research Engine (HURE). The engine was tested in the PSL with ice crystal cloud ingestion over a range of simulated altitudes and operating conditions. The test took place in January 2018, as part of the engine icing research supported by the NASA Advanced Aircraft Icing Subproject, under the NASA Advanced Air Transport Technology Project. Since the HURE was never in production, it has not had any inflight events that have been attributed to ice crystal ingestion. The engine was installed into the PSL altitude facility with a direct connect duct thatmates to the engine flange, such that it utilizes the full flow capacity of the PSL altitude wind tunnel as illustrated inFigure . There was no flight nacelle

10 installed in this test configuration. T
installed in this test configuration. The PSL is an altitude engine testing facility located at NASA Glenn Research Center. The PSL features the Escort data acquisition system. Each engine datapoint is stored in the system and is referred to as an Escort data point number. The Escort system records the data at a frequency of approximately one scan per second. ��NASA/TM220023Figure PSL direct connect inlet duct connects thePSL altitude test facility to engine flangeFigure Fullscale engine testing in 3 at simulated altitude with ice crystal cloud ingestion.The spray bars areshown in the left photo, and the direct connect piping are in the right photo.This test facility features water spray bars that can produce a fully glaciated ice crystal cloud with controllable ice particle size and concentration per unit volume of air.The spray bars (left photo) and the directconnect inlet duct piping are shown below in Figure with the engine and test stand (right photo).The purpose of the testwas to determine if ice would accrete within the compressor components at the predicted operating conditions outlined in Reference 8. The HURE engine (crosssection illustrated inFigure ) was heavily instrumented with traditional pressure and temperature gauges, as well as video cameras at key locations withinthe compression system. Prior to the test, an extensive study utilizing a computational tool was performed to determine the operating conditions where a risk of ice accretion would be expected (Ref. 8). The study resulted in numerous operating conditions and altitudes where the ice would be expected to form, at several locati

11 ons within the compression system. The c
ons within the compression system. The computational tool that was utilized consisted of an enhanced version of the COMDES mean line compressor flow code (Ref. 9). This tool was previously applied to analyze icing data from other engines tested in PSL with ice crystal ingestion (Refs. 10 to 16). That research resulted in determining values for key parameters that can indicate whether there is a risk of ice accretion. These parameters form the basis of the Icing Wedge. The Icing Wedge is defined by thresholds of static wet bulb temperature, icwater flow rate to air flow rate, and a nonzero particle melt ratio. Leveraging from the ��NASA/TM220023Figure Honeywell Uncertified Research Engine (Courtesy Honeywell Engines). previous analytical studies, it was hypothesized in Reference 8 that the Icing Wedge is universally applicableto other turbofan engines. The primary focus of the study in Reference 8 was to determine the engine operating conditions and ambient temperatures at various altitudes, that would result in ice accretion between the fanstator, splitterlip, and the inletguide vane (IGV) of the highpressure compressor (HPC). The secondary focus of that study was to enableaccretion to occur in the variable stator of the HPC first stage (tator 1). These targeted areas where the ice was expected to occur and the associated station numbers from the compressor flow simulation are illustrated in Figure The current study is focused on the analysis of 57 Escort test points that were taken at distinct operating conditions in PSL at altitudes between 5K and 45K ft. The computer analyses results for these data poi

12 nts with the Customer Deck as well as wi
nts with the Customer Deck as well as with the COMDESMELT codes are listed in the Appendices A, B, and in Reference 17. The results of the HURE testing confirmed by the video that ice accreted and, or collected at all altitudes in the general targeted regions of the fanstator, splitterlip, and the IGVconfirming most of the predictions made in Reference 8. During testing, it was observed that ice accreted at the fanstator in accordance with the icing parameter thresholds governed by the Icing Wedge. Note that this fanrotor is highly loaded, causing a significant rise in total temperature, which is adequate to partially melt the incoming ice particles.Ice accretion was observedon the front frame components near the splitterlip at lower static wet bulb temperatures than was expectedbased on the Icing Wedge minimum thresholdof 492. In addition, the accretion occurred at static air temperatures well below freezing, thus noparticle melting could have occurred due toheating from the air alone. The aluminum front frame may have received heat from additional sources besides the air, but this process is not well understood for this engine. It is possible that the ice accretion in that region was not an adiabatic process. During testing, in order to compensate for the lack of a heat transfer model in the COMDESMELT code, the target static wet bulb temperature (Twbs) for ice to accrete in the front frame region was reduced to 468R. This was 24R below the Icing Wedge minimum threshold of 492R. The new target Twbs was determined by the analysis of one of the operating points where ice accreted at the splitterlip and shroud region. U

13 sing COMDESMELT, new testing conditions
sing COMDESMELT, new testing conditions were rapidly derived prior to further testing, and the test matrixwas modified. This was successful in enabling iceto accrete in the front frame components (splitterlip and shroud region).For test data analysis, a simple bulk heat transfer model was developed to estimatethe wall metal surface temperature. This was done in order to compare it to previous enginetests which had measured wall temperatures between 493to 501R during ice accretion. ��NASA/TM220023Figure The targeted locations where the ice was expected to occur are the fanstatorplitterlip through IGV, and the HPC Stator 1. Also shown are the associated station numbers used in the compressor flow simulation.Video observation indicated that ice collected on the variable inlet guide vane (IGV) of the highpressure compressor. This may have been due to partially melted particles flowing along the shroud wall from upstream, that collected on the IGV surface. This appeared to be different from the ice accretion observed at the upstream regions. Due to the nonadiabatic process upstream of the IGV, the Icing Wedge thresholds, as an indicator of accretion risk, are not applicable in this regionThe test matrix also included three operating conditions where ice was expected to accrete on the HPC variable tator 1, based on the pretest analyses. However, during testing it appeared that ice did not accrete there, as evidenced by the measured metal temperatures of the variable tator 1There were no video cameras in this region to verify whether ice accreted. Potentiallythe additional heat source to the particles from the

14 upstream shroud wall near the splitterli
upstream shroud wall near the splitterlip melted the particles beforeentering the HPC. he energy imparted by HPC rotor 1 further increased the air temperature, thus raising the mixedphase “icewater” temperature entering tator 1. The testing of the HURE indicatedthat ice accreted or collected at all altitudes tested and was confirmed by video cameras at the three targeted locations of fanstator, splitterstrut, and IGV. Note that the accretion in those locations occurred at different operating conditions. ose test points are superimposed onto the historical engine icing events reported in References 1 and 2, as illustrated inFigure . The historical icing events occurred on commercial airlines (Figure ) during flight through clouds with high icewater content. Those events have been attributed to ice crystal ingestion and subsequent ice buildup in the engine, but the exact location within the compression system is unknown. The plot in Figure includes reference lines indicating the International Standard Atmosphere (ISA) temperature, with +18F and+36F above ISA, versus altitude. Ice accretion within the HURE engine spanned the entire range of the reported historical icing events for commercial engines. The PSL test conditions at 5K ft altitude were set to inlet temperatures on the order of F, which are offset by F from the ISA temperature, in order to induce ice accretion in the targeted location of the compression system. Note that this condition does not occur in nature, but was artificially set in PSL for icing research and code modelingdevelopment ��NASA/TM220023Figure The confirmed operating poi

15 nts with ice accretion or collection in
nts with ice accretion or collection in the HURE engine during PSL testing, superimposed onto the reported commercial engine icing events (Ref. 1).Posttest analysis showed differences between the engine thermodynamic system model and the measured engine performance data. These comparisons between the system model and the tested engine performance were based on data prior to icecloud ingestion. These variances that were on the order of 6% may be partially responsible for uncertainties in the posttest analyses with the COMDESMELT code, which depends on the engine system model as well as the measured test data for boundary conditions. Compressor Code CalibrationPrior to testing in PSL, the COMDES code was compared to the design point performance results from a customer deck, as well as more detailed results from a twodimensional streamlinecurvature analysis code for the fan and the four axial compressor stages of the HPC. These two results were provided by Honeywell for the purpose of calibrating the COMDES code atthe design point. These are listed in Appendix A of Reference 17, and include the summary of the HURE fan and HPC geometry, as well as the streamlinecurvature analysis results, and the COMDES flow code results at each blade edge.After testing the HURE engine, in order to further verify the accuracy of the COMDES compressor code in being able to model the flow through the full fan and fanre region, six test data pointswithout ice cloud ingestion were analyzed with the COMDES code(Appendix B of Ref. 1The purpose was to confirm that the compressor aerodynamic performance predicted by the flow model matched the meas

16 ured aerodynamic performance test data o
ured aerodynamic performance test data of the compressor within the engine. In addition to calibrating the rotor and stator losses withinthe flow model, the effects of the aerodynamic instrumentation installed in the compression system could also potentially influence the measured performance. The effect of the ��NASA/TM220023instrumentation was not significant, since the measured and the fan and compressor aero performance (total temperature and pressure, and static pressure) modeled with the CD and COMDES were all in good general agreement (see Figures 42 to in Appendix Bof Ref. 17). Therefore there was no compressor code calibration required to accurately model the fan, flow splitter, and IGV regions. During testing even in an ice cloud environment, it was observed that the total pressure and the wall static pressure measurements in the fanstator and core strut regions remained unaffected by the particles. However, the cascading aerodynamic effects of small differences in the fan performance can propagate through the HPC and affect the compressor exit conditions (Pt and Tt) through the stage matching. This effect may be responsible for the observed differences in engine performance parameters of up to 6percentbetween the CD engine system model and measured Escort test data. Even though the compressor exit plane is well downstream of any ice accretion sites, the differences in conditions there may influencethe calculation of core mass flow in the CD system model. FanStator Analysis; Risk of Accretion Governed by theIcing WedgeThresholdsThe analysis of the HURE test data was performed with the similar c

17 omputationalprocess shown in Figure (a)
omputationalprocess shown in Figure (a) thatwas utilized in the predictions of icing risk of Reference 8. The computational process is further detailed in Appendix Cof Referenc, with the description of the input parameters for theengine thermodynamic cycle codeas well as for the COMDESMELT codeInitially the CD is executed at the tested altitude, ambient temperature, flight Mach number, and fan speed. The results from the engine CD model are utilized as boundary conditions to the subsequent execution of the compressor flow analysis code (COMDESMELT). The compressor flow analysis is performed to determine the aerodynamic flow field as well as the thermodynamic state of the ice particle, in addition to computing the key icing parameters of wet bulb temperature and IWAR. These analyses assumed that the ice particles would be distributed uniformly through the bypass as well as the engine core, even though this engine hasa “hidden core.” As part of the ice particle thermodynamic state, the melt ratio is calculated to determine the existence of liquid water content, which is a requirement for there to be ice accretion. Since there is no particle break up model in the code,the analyses was performed over a range of possible particle sizesfrom 3 10 The calculated icing parameters are then compared to the minimum and maximum threshold limits of the Icing Wedge shown in Figure (b), and thevideo images of the compressor components are evaluatedto determine if ice accreted on the component. Note that the threeimensional Icing Wedge shown inFigure (b)was derived from a previous turbofan engine test, where the blockage grow

18 th rate due to ice accretion was iterati
th rate due to ice accretion was iteratively determined to match the measured wall staticpressure and the total pressure ratioThis study only utilizes the minimum and maximum threshold values of the static wet bulb temperature (Twbs) of the Icing Wedge that indicate the risk of accretion on the fanstator. Estimates of blockage growth rate due to accretion and boundary layer growth are not part of this current study.The values of the static wet bulb temperature (Twbs) in the fanstator are utilized as a verification of accuracy for predicting the risk of ice accretion. If the value of Twbs falls between the Icing Wedge thresholds, then there is a risk of ice accretion. The static wet bulb temperature is calculated from the local values of static air temperature and the relative humidity.The location of ice accretion on the fanstator was observed in the video to be at discrete radial locations at each operating condition (Escort test point), but for some test points the accretion appeared to be at a higher span than the notional streamline of the fancore. It became apparent that a full fan flow analysis was also required to determine the Twbs at the higher RMS radius of the full fan. ��NASA/TM220023 (a) (b) Figure (a)Computational process for the test data analysis for the fan stage of the HURE turbofan engine.(b) The Icing Wedge with the thresholds of Twbs between 492 498R for a risk of accretion.For this reason two models were created of the fanstator. The first model was of the full fanrotor and fanstator. This analysis resulted in computing the aerodynamic parameters at the rootmeansquare (RMS)

19 radius of the full fan stage, as illustr
radius of the full fan stage, as illustrated in Figure . The computed parameters included: relative and absolute velocities, static and total pressures and temperatures, absolute and relative flow angles, including the static wet bulb temperature and the particle melt ratio at the leading and trailing edges. The second model was of the fancore region through the splitterlip, and the variable IGV. In this analysis, a notional stream line was assumed that divided the flow between the bypass and thecore, as illustrated iFigure The notional streamline was used as the fancore flow pathouter wall. Likewise that analysis resulted in modeling all the aerodynamic parameters, including the static wet bulb temperature (Twbs) and the particle melt ratio at the RMS radius of the fancore stage, through the IGV. The computed Twbs and IWAR of the Escort test points analyzed in the fan stator region are illustrated in Figure . The symbols connected by dotted lines represent the values of Twbs at the full fan and at the fancore RMS radii. The spanwise work distribution in the fan rotor resulted in a gradient of static temperature, resulting in a large variation of Twbs. The minimum and maximum Twbs thresholds (492 to R) of the Icing Wedge from Reference 16 are represented by two horizontal lines. These are superimposed onto the calculated values of Twbs obtained from the HURE test data analyses. Figure represents the static wet bulb temperature at Station 5 shown iFigure The black symbols in Figure indicate the Escort test points where ice accretion on the fan stator was confirmed by video observation. For most of the test points that e

20 xperienced accretion, the range of Twbss
xperienced accretion, the range of Twbsspanned the minimum and maximum thresholds of the Icing Wedge. The combined analysis of the fancore and full fan results had a 94 percent success rate of spanning the Icing Wedge for all data points analyzed which had ice accretion in the fanstator. e test data points where ice accretion was not observed by video in the fanstator, were outside the Icing Wedge thresholds, as indicated by the red symbols in Figure . The analysis results had a 100 percent success rate of being outside the Icing Wedge for all data points analyzed which had no ice accretion in the fanstator. ��NASA/TM220023Figure Two flow models of the fan stage: (a) Full fan; (bFancore, with its outer flow path wall indicated by the notional streamline.The numbers 1 6 refer to the meridional stations.Figure Range of static wet bulb temperature at the fancore stator and full fanstator rootmeansquare (RMS) radii. ��NASA/TM220023In summary, there was good agreement between the Twbs thresholds of the Icing Wedge and the analysis of the HURE test results. However, it was necessary to have both the full fan and the fancore models to determine the range ofTwbs along the full span of the fanstator.The static wet bulb temperature is based on accurate calculations of the local static air temperature as well as the local relative humidity within the flow field of the fan as well as the four axial stages in the core. A sample data point is shown iFigure , where the inlet relative humidity is 100percent, illustrates the rapid drop in relative humidity. The specific humidity increases due to the sublimat

21 ion, melting, and evaporation of the ice
ion, melting, and evaporation of the iceparticles. The accretion at tation 5 (fanstator trailing edge) occurs at approximately percentrelative humidity as shown in Figure The operating conditions for the Escort data points where there was no ice accretion in the fanstator are listed in Table . All of the data points with no ice accretion were at static wet bulb temperatures below the minimum threshold of 492R, except for one (Escort point 212), which was above the Icing Wedge maximum threshold temperature of 498R. Escort is the data recording system in PSL. It records the data at a frequency of approximately one scan per second. The computer analyses results for these data points with the Customer Deck as well as with the COMDESMELT codes are included in Appendix C and listed by Escort number.Figure Distribution of relative and specific humidity through the compression system. ��NASA/TM220023TABLE FANSTATOR: NO ICE ACCRETION Escort ata oint Altitude, K ft Flight Mach Number Ambient Temp ., °R Delta ISA, °R Tt PSL Station 1 ce loud Pt PSL Station 1, psia Ice Water Content (IWC) IWAR Twbs an ore & ull Stator, Station 5, 5 M elt ratio, Station 5, 5 147400.62397.87.9428.53.491.280.0041457.4 468.90.0 0.0 2440.19429.871.4433.112.6110.580.0089462.8 461.70.0 0.0 300450.77388.31.6434.23.112.950.0102453.1 453.30.0 0.0 2500.20432.868.3436.212.615.900.0052472.5 478.60.0 0.0 2420.21432.968.0436.912.6410.500.0092468.1 464.20.0 0.0 149400.61406.917.0437.73.491.540.0049461.8 465.70.0 0.0 2450.20434.366.6437.912.6012.000.0101468.4 465.50.0 0.0 2460.20434.866.2438.412.616.200.0052468.7 465.70.0 0.0

22 299450.77392.82.9439.43.122.920.0102456.
299450.77392.82.9439.43.122.920.0102456.7 456.90.0 0.0 279250.60410.119.6439.76.985.800.0089464.7 456.20.0 0.0 280250.60411.118.5441.06.985.800.0089465.6 457.00.0 0.0 282360.61412.222.7442.64.211.000.0026463.4 456.80.0 0.0 287360.61412.222.3443.14.222.480.0066467.3 484.50.0 0.0 283360.61412.722.8443.34.212.000.0051464.2 457.30.0 0.0 284360.61412.822.9443.44.214.100.0105464.9 457.90.0 0.0 289360.61413.523.4443.94.212.410.0061459.7 446.00.0 0.0 2430.20440.560.6444.012.6110.600.0091473.7 470.50.0 0.0 298450.78396.26.4444.23.142.950.0103459.9 463.30.0 0.0 293450.77402.812.9450.53.122.280.0081463.3 461.20.0 0.0 295450.77403.113.2450.53.132.950.0102461.5 449.30.0 0.0 156450.77403.613.7451.03.132.730.0096465.5 468.90.0 0.0 291450.77403.613.7451.43.120.750.0027463.0 461.10.0 0.0 292450.77403.113.2451.53.121.330.0047463.5 461.40.0 0.0 153450.81399.49.5452.23.282.890.0099470.2 470.50.0 0.0 157360.61422.332.4453.64.181.460.0039471.4 467.50.0 0.0 297450.76407.217.3454.93.122.930.0105466.4 464.60.0 0.0 196300.81427.015.8482.56.651.920.0034489.3 476.50.0 0.0 212250.80429.10.4484.58.342.600.004498.6 510.30.20 0.41 ��NASA/TM220023Figure shows the view of the fanstator from the video camera mounted on the engine shroud facing radially inward towards the hub. At Escort reading 287 no ice accretion is observed in the fanstator, and is typical of all the test data points listed in Table The image shown inFigure , is a typical case (Escort 121) where the ice accretion was observed on the fanstator through the video cameras. All the Escort data points which experienced ice accretion on the fanstator are listed in Table . E

23 ight Escort data points at the bottom of
ight Escort data points at the bottom of Table (103209) had significantly less ice accretion than the other data points in the table.Figure Video camera view of the fanstator showing that no ice accretion occurred there during Escort data 287.Figure Video camera view of the fanstator showing that ice accretion occurred thereduring Escort data 121. ��NASA/TM220023TABLE FANSTATOR: ICE ACCRETIO Escort ata oint Altitude, K ft Flight Mach Number Ambient emp Delta ISA, R Tt PSL Station 1 ce loud ON, °R Pt PSL Station 1, psia Ice Water Content (IWC) IWAR Twbs an ore & ull tator, Station 5, 5 Melt fatio, Station 5, 5 100250.40436.710.4450.95.862.930.0061481.9 505.00.0 0.28 276250.41436.87.3451.36.122.930.0059481.4 505.40.0 0.28 187250.41437.57.9451.96.122.930.0059482.4 505.50.0 0.283 193300.61429.318.2461.15.562.890.0065483.5 504.40.0 0.27 122300.60429.918.4461.25.572.880.0065484.4 505.10.0 0.0 191300.61429.918.7461.75.572.900.0066483.8 504.50.0 0.271 188300.60431.119.8462.75.572.840.0064485.7 504.80.0 0.283 214250.81412.416.9466.08.344.000.006487.5 503.70.006 0.17 123400.81414.524.6469.34.082.900.0089483.7 498.10.0 0.107 124350.81420.327.3475.25.252.890.007489.1 504.00.019 0.272 108350.80424.831.7479.95.251.520.0036490.4 499.10.03 0.114 114400.81428.338.4484.24.091.420.0044489.2 500.90.02 0.22 121450.81427.938.0484.43.281.480.0058487.0 498.50.0 0.16 118400.82427.737.9485.74.081.460.0043488.1487.40.0 0.0 103250.65434.75.0471.77.272.500.0042491.7 493.40.037 0.002 105250.65434.54.9471.87.282.360.0037487.6 480.10.0 0.0 213250.80418.311.3472.38.352.600.004491.0 505.80.05 0.24 159450.76424.6

24 34.7473.63.120.500.002481.0 505.20.0 0.0
34.7473.63.120.500.002481.0 505.20.0 0.0 110300.81426.014.8481.66.661.950.0027492.3 488.00.053 0.00 112300.80426.715.1481.76.662.200.0038489.1 476.60.0 0.0 208400.83427.337.4485.64.081.000.0031489.2 500.60.021 0.22 209400.80430.640.7486.14.080.800.0025489.5 500.60.028 0.22 ��NASA/TM220023Figure illustrates the static wet bulb temperature distribution through the full fan (represented by blue lines), and the fancore and the four stage axial compressor (represented by black lines), as well as the particle melt ratio over a range of particle size from 3 to The computer analyses results for these data points listed in Table areincluded in Appendixof Reference 17, listed by Escort numberTable and Table show the static wet bulb temperature and the corresponding particle melt ratio for a 5 particle at the trailing edge of the fanstator (Station 5 inFigure and Figure ). However a range of particle sizes were analyzed with the COMDESMELT code between 3 and 10 forall the stations in the flow path. The full fan model had 5 meridional stations, while the fancore model is comprised of 28 meridional stations, as shown in Figure , since the four axial stages of the high pressure compressor were also included as part of the flow analysis. The calculated Twbs downstream of the HPC rotor 1 (meridional station 9) was significantly higher than the maximum threshold of the Icing Wedge. Therefore there was no risk of accretion in the downstream stages, although there were no video cameras in those stages for visual confirmation. For the Escort data points where ice accretion occurred on the fanstator, there was no accretio

25 n observed concurrently on the splitterl
n observed concurrently on the splitterlip, shroud, strut, IGV regions. Likewise, when ice accretion was observed in the splitterlip, shroud, strut, IGV regions, there was no accretion observed concurrently on the fanstator.Figure Ice accretion occurred in the fanstator during Escort 121. The static wet bulb temperature was within the Icing Wedge. ��NASA/TM220023SplitterStrutGooseneck Region; NonAdiabatic WallPSL Test, and RealTime Predictions ofIce Accretion Operating PointsEarly in the HURE testing, it was observedthat at several of the tested operating points, ice accretion in the front frame region (splitter, shroud, gooseneck and support strut) did not occur as expected even though the static wet bulb temperature(Twbs) were within the thresholds of the Icing Wedge. Although there were no metal thermocouples to measure the wall temperature near the aluminum splittershroud region, it is assumed that the metal temperatures were above freezing, sinceliquid water streaks were observed in the videos on the shroud. The target static wet bulb temperature was reduced by 24R from the Icing Wedge minimum threshold value of 492in order toaccrete ice in the region of the front frameThe value of 468was determined through flow modeling in near realtime analysis of several test data points that experienced ice accretion at the splittershroud regionlow analysis of those data pointsindicated that the calculated static air temperature was well below freezing at this location. Thus no melting uld have possibly occurred solely due tothe heat of compression caused bythe fan. In addition, the calculated temperature of the

26 ice particles was also well below freez
ice particles was also well below freezing, based on the flow analysis. New testing operating conditions over a wide range of simulated altitudes were rapidly determined utilizing the lowerTwbs threshold in the splittershroud region prior to further testing, and the test plan was modified accordingly. With these updated conditions ice began to accrete within seconds after the ice cloud was initiated. It was assumed that the splittershroud region wall temperatures at this condition were near freezing, since the measured metal temperatures on the IGV were at freezing. This technique was successful in inducingice to accrete in the splittershroud region, for most subsequent tested operating points. Note that when ice accretion was observed in the splitterstrut, shroudregionthere was no accretion observed on the fanstator at the same operating points.he aluminum front frame may have received heat from additional sources besides the air, in order to partially melt the particles, which then subsequently accreted near the splittershroud. A physicsbased explanation was sought for this phenomenon. It is possible that the ice accretion in thregion is due to a adiabatic wallAlthough the source of additional heat flux isnot completely understood, a simpleordermagnitude heat flux model was implemented in COMDESMELT to estimate the wall temperature this region. OrderMagnitue Estimate of Wall TemperatureIn order to reconcile the additional enthalpy from the nonadiabatic flow path wall, an ordermagnitude method was implemented that assumes a continuous supply of heat which is transferred directly to the ice particle by conduction.The

27 heat transfer model was calibrated empir
heat transfer model was calibrated empirically based on the HURE test results. Since there were no thermocouples in the splittershroud region of the front frame where ice accreted, it was necessary to utilize measured IGV metal temperature test data to develop the heat transfer model for the front frame.The purpose of the heat transfer model was to calculate the wall temperature in the splittershroud region. Several data points were observed to have glassy ice accretion at the splitterlip and shroud of thegooseneck with no shedding (hard ice). At these operating points it was observed that the measured metal temperatures at the IGV were near 492R. It was therefore assumed that the metal temperature at the splitterlip and shroud were likewise near 492R at these operating conditions. Table below lists four test data points at various altitudes that were used to confirm that when hard ice was observed to accrete on the splitterlip and (or) the shroud, the IGV had a measured metal temperature near R. This was also observed in the previous engne test (Ref. 15 see Figures 14 and). The total air temperature at the splittershroud region was approximately the same in the CD and COMDES models, and confirmed the measured Escort data. The calculated static air temperature in the splittershroud region as modeled by COMDES was consistently below freezing. Yet it was observed that ice accreted in the splitterstrutshroud region. ��NASA/TM220023TABLE DATA POINTS TILIZED TO CALIBRATEHEAT TRANSFER MODEL IN COMDESMELT Test d ata p oint scort) Altitude, K ft Splitter - s hroud totalir, CD, scort andCOMDES ° R Splitt

28 er - s hroud staticir CD,COMDES, Tigv
er - s hroud staticir CD,COMDES, Tigv m etal re cloud, Tigv m etal ost loud;after 60 sec, 156 45 509.4 486.3 511.2 492.3 283 36 499.9 481.9 501. 1 491.0 279 25 496.1 478.1 494.2 491.9 242 5 492.8 476.9 492.0 491.9 A simple heat transfer model was created, where the initial unknown was the bulk heat transfer coefficient at the splittershroud flow path wall, wall plenum. To estimatewall plenum, two of the test data points were selected (156 and 283) which had visual confirmation of glassy, hard ice accretion which did not shed, on the splitterlip and shroud. Although the exact source of heat (enthalpy) to the walls was not known, in this study it is assumed that the heated air from a downstream stage was the source. In order to have this additional enthalpy vary with engine operating conditions, for modeling purposes it was assumed that the source was the HPC stage 2 exit (COMDES; Tt2). In the analysis that follows, the temperature of the plenum, wall plenum, wasassumed to be the same Tt2. At the selected data points (156 and 283), the heat flux, was calculated assuming that the splitter andshroud wall temperatures, wall core, were at 492R with the following quation (1), where the is the mass flow rate of the ice particles going through the engine core. ����������������

29 DC5D;(1)The mass flow of ice particles t
DC5D;(1)The mass flow of ice particles through the engine core is assumed to be proportional to the air flow rate, since the compressor code lacked a model to track the particles within the flow path in threedimensional space. The SHCis the specific heat capacity of the ice particles. The ice particle temperature was obtained from an initial solution of the COMDESMELT code. Knowing the heat flux, the temperature on the plenum side of the shroud wall, wall plenum, was computed with Equation (2)��������� ������ �����(2) The in the above equation is the surface area of the splittershroud region through the entire goose neck, while is the coefficient of conduction for aluminum, and thickness is the average wall thickness of the aluminum shroud casing. The bulk heat transfercoefficient, wall plenumwas determined for this engine based on the assumption of wall temperature of 492R with quation (3). ������������������

30 ��
����� (3) For the flow analysis of all other data points, the bulk heat transfer coefficient was assumed to remain at the value which was calculated from the data points 156 and 283 of Table . The above three equations were incorporated into COMDESMELT and are solved for the three unknown parameters (Q, Twall plenumwallcore) for all subsequent datapoints analyzed in this study.COMDESMELT was executed iteratively to first, determine wall plenum(Tt2)and particle temperature at the splittershroud region, and finally ��NASA/TM220023Figure Computational Process for NonAdiabatic Walls that was incorporated into the COMDESMELT code to estimate the core watemperature at the splittershroud region.executed to calculate the splittershroud wall temperature (wall core). However, in the current study the enthalpy extracted from the heated wall was not transferred back to the ice particle to determine its thermodynamic state once it made contact with the heated splittershroud wall. The computational process summarizing the steps above is outlined in Figure and in Appendix C of Reference 17The computed value of wallcorewas then compared to previous engine test data (HoneywellALF502R5, serial number LF11) with measured wall metal temperatures at the location where there was significant ice accretion (Ref. 15) and are illustrated inFigure . Note that additional analysis of the engine data in Reference15 was performed as part of this study, and is illustratedFigure . Ice accretion occurred on the LF11 compressor shroud wall at me

31 asured metal temperatures on the shroud
asured metal temperatures on the shroud wall between 493 and501R. The wall corecalculations for all HURE test data points which were analyzed are illustrated inFigure . The range of calculated metal wall temperatures at the splittershroud region using the heat transfer model described above where ice accretion was confirmed by video, were between 475 to 51R, and span the range of wall temperatures measured in the LF11 engine during accretion. The calculated HURE wall temperatures were superimposed onto the measured wall temperatures of the LF11 engine inFigure . The figure illustrates LF11 metal temperatures that supported accretion, as well as a few data points where accretion was not observed.During the HURE testing it was observed through video confirmation that there were two distinct types of ice accretion at the splittershroud region, which did not shed. These are what were defined as “ice accretion” and appeared to be glassy and fully transparent, and are plotted iFigure and Figure withsolid blue circles. An example of each of these two types of ice accretion is highlighted in the following section.The first type of distinct ice accretion occurred on the splitterlip at discrete circumferential locations, and often had protruding horn shapes growing upstream, as shown inFigure ��NASA/TM220023Figure Honeywell ALF502, serial LF11 measured wall temperatures with video confirmed ice accretion. The closed symbols resulted in engine rollback, while the open symbolshad ice but did not result in engine rollback. (Ref. 15)Figure Calculated wall metal temperature of the HURE splittershroud region,

32 compared to themeasured wall temperature
compared to themeasured wall temperature of the LF11 engine (Ref. 15), versus IWAR. ��NASA/TM220023Figure Calculated wall temperature versus IWAR of the HURE splittershroud region for data points with ice accretion at high altitudes and at 5K ft, compared to the measured wall temperature of the LF11 engine (Ref. 15).Figure Escort 153. Ice accretion with ice horns on the splitter. ��NASA/TM220023Figure Ice accretion occurred at the splitter shroud during Escort 153.The static wet bulbtemperature is 19R below the Icing Wedge minimum threshold of 492For Escort data point 153, the compressor flow model was executed to determine the static wet bulb temperature distribution through the full fan, the fancore and the four stage axial compressor, as well as the particle melt ratio over a range of particle size for 3 10 , and is illustrated in Figure . The results of the flow analysis for this data point indicated that the static wet bulb temperature at the splitteshroud region was 19R below the Icing Wedge minimum threshold.Escort data point 156 is an example of the second distinct type of ice accretion in the splittershroud region in which ice accreted on the shroud surface but without ice horns on the splitter, and is shown inFigure . For Escort data point 156, the compressor flow model was executed to determine the static wet bulb temperature distribution through the full fan, the fancore and the four stage axial compressor, as well as the particle melt ratio over a range of particle size for 3 10 , and is illustrated in Figure . The flow model indicated that the splittershroud surface wet bulb te

33 mperature was 24R below the Icing Wedge
mperature was 24R below the Icing Wedge minimum threshold of 492 ��NASA/TM220023Figure For Escort 156, there was ice accretion on the shroud, without horns.Figure Ice accretion occurred at Escort 156. The static wet bulb temperature is 24R below the Icing Wedge minimum threshold of ��NASA/TM220023For all data points experiencing ice accretion, key aerodynamic andicing parameters are listed inTable . The computer analyses results for these Escort data points with the Customer Deck as well as with the COMDESMELT codes are included iAppendix of Referenceand listed by Escort number.TABLE SPLITTERSHROUD: ICE ACCRETION DATA POINTS Escort data point Altitude, K ft Flight Mach Ambient temp ., °R Delta ISA, °R Tt PSL Station 1 ice cloud ON, °R Pt PSL Station1, psia Ice Water Content (IWC) IWAR Twbs splitter, shroud, fan c ore (6), 5 Ice m ass flow into core, l bm/sec Kinetic energy heat flux, Qk Conduct eat lux, Qc, BTU/Hr Total eat lux, Qt, BTU/Hr T wall core after 60s (6) 2.1, T ice particle (6) 2.1, °R 147400.62397.87.9428.53.491.280.0041462.50.0470.29236263626501.3458.3 30045 0.77 388.31.6434.23.112.950.0102456.60.0840.40432503251474.9453.5 149400.61406.917.0437.73.491.540.0049465.50.0490.26332893289499.2462.2 299450.77392.82.9439.43.122.920.0102460.10.0850.41732673267478.4457.0 279250.60410.119.6439.76.985.800.0089467.90.1550.65028182818475.3465.2 280250.60411.118.5441.06.985.800.0089468.80.1530.64528662867476.5466.1 282360.61412.222.7442.64.211.000.0026466.00.0260.10420702070507.9463.5 287360.61412.222.3443.14.222.480.0066471.50.0850.49838453846493.1

34 467.9 283360.61412.722.8443.34.212.000.0
467.9 283360.61412.722.8443.34.212.000.0051466.80.0520.20726182618492.4464.3 284360.61412.822.9443.44.214.100.0105467.50.1060.42629972997480.7465.1 298450.78396.26.4444.23.142.950.0103463.10.0840.40733903390482.5460.2 293450.77402.812.9450.53.122.280.0081466.20.0640.29731093109490.7463.5 295450.77403.113.2450.53.132.950.0102463.20.0630.20023402340481.8461.3 156450.77403.613.7451.03.132.730.0096467.90.0730.31432733273492.0467.1 292450.77403.113.2451.53.121.330.0047466.30.0370.17426512651503.3463.6 153450.81399.49.5452.23.282.890.0099473.10.0880.44139953995495.5470.3 157360.61422.332.4453.64.181.460.0039474.10.0410.18229282928511.6471.6 297450.76407.217.3454.93.122.930.0105469.20.0820.39132863286488.7466.6 ��NASA/TM220023The range of static wet bulb temperatures of the Icing Wedge for these ice accretion points was from to 474.1R with a 5 particle in the flow analysis model. The average difference of the Twbs for these ice accretion points was 25R below the 492R minimum threshold Twbs of the Icing Wedge, with a range of 17.5see Figure Note that due to the additional heat from the splittershroud region, the Twbs threshold for hard ice accretion is significantly below the 492R of the Icing Wedge for this engine.It was also observed through video that on many test data points ice would collect on the splitterlip but would rapidly shed at up to 3 sheds per second. This type of ice was not glassy, but rather appeared to be white and not transparent. These data points are referred to as “ice collection” data points in this study, since they did not appear to adhere to the surface, but rather the ic

35 ewater particles appeared to have been m
ewater particles appeared to have been merely collected at the splitterlip stagnation point, and rapidly shed. These data points are illustratedas open pink circles in Figure andFigure . As evident from Figure these data points with rapid shedding occurred at conditions that were at warm metal temperatures, as calculated with the heat transfer model in the compressor code.A typical data point featuring ice collection on the splitterlip with rapid shedding is shown inFigure he ice collection appear as small white ice “mounds” which were quite different from the glassy hard ice noted in the previous ice accretion cases. These mounds appeared at circumferentially discrete locations on the splitter lip, and shed at a rapid rate of from multiple times per second to several seconds per shed. The calculated metal temperatures for these data points were in the range from 505.4 to 563.9R. These were well above the maximum measured wall temperatures of 501R that supported ice accretion from Reference 15.Figure Calculated wall temperatureversus IWAR of the HURE splittershroud region with no ice accretion, only ice collection and rapid shedding. Black symbols are the measured wall temperaturesat the location of ice accretion (Ref.15). ��NASA/TM220023Figure Ice collection on the splitterlip with rapid shedding, is not considered ice accretion since it does not grow (Escort 276).Figure Ice accretion did not occur for Escort point 276 due to additional heat from the warm wall at the statorshroud region. The static wet bulb temperature there was on the order of 487.9Figure illustrates the static wet bulb temper

36 ature distribution for Escort points 276
ature distribution for Escort points 276 through the full fan, the fancore and the four stage axial compressor, as well as the particle melt ratio over a range of particle size for 3 . Ice accretion did not occur at the statorshroud region due to the additional heat from the splittshroud wall, as the wet bulb temperature was 5R below the Icing Wedge minimum threshold limit. For data points that experienced ice collection with rapid shedding, but no ice accretion, the key aerodynamic and icing parameters are listed in Table . The full analysis results for these Escort data points with the Customer Deck as well as with the COMDESMELT codes are included in Appendix C of Referenceand listed by Escort number. ��NASA/TM220023TABLE SPLITTERSHROUD:ICE COLLECTION WITH RAPID SHEDDING; NO ACCRETION(CALCULATED HOT WALL Escort data p oint Altitude, K ft Flight Mach Ambient t emp., Delta ISA, °R Tt PSL Station 1 ice c loud ON, °R Pt PSL Station 1, psia Ice Water Content (IWC) IWAR Twbs splitter, shroud, fan core (6), 5 Ice mass flow into c ore, bm/sec Kinetic energy heat f lux, Conduct heat f lux, Qc, BTU/Hr Total heat f lux, Qt, BTU/Hr T wall c ore after 60s (6) 2.1, °R T ice particle (6) 2.1, °R 100250.40436.710.4450.95.862.930.0061487.90.1331.02955245525506.3483.2 276250.41436.87.3451.36.122.930.0059487.90.1351.08354575458505.4482.9 291450.77403.613.7451.43.120.750.0027465.90.0210.09821162116.1519.2463.1 187250.41437.57.9451.96.122.930.0059488.60.1341.04956125614507.1483.7 193300.61429.318.2461.15.562.890.0065489.90.1331.11955735574508.2485.0 122300.60429.918.4461.25.572.880.0

37 065490.50.1311.06956085609509.5485.7 191
065490.50.1311.06956085609509.5485.7 191300.61429.918.7461.75.572.900.0066490.20.1341.11855655566508.6485.4 1880.60431.119.8462.75.572.840.0064491.50.1281.02556615662511.5486.9 214250.81412.416.9466.08.344.000.006494.20.1741.40753995400506.4489.1 1230.81414.524.6469.34.082.900.0089488.60.1250.96956765677509.9484.6 1030.65434.75.0471.77.272.500.0042497.60.0830.46741934193519.7491.7 105250.65434.54.9471.87.282.360.0037490.20.0650.28340994099522.5487.9 213250.80418.311.3472.38.352.600.004497.20.1120.90052475248517.6491.5 1240.81420.327.3475.25.252.890.007494.40.1271.02456825683515.0490.1 108350.80424.831.7479.95.251.520.0036494.60.0590.42045444544533.3490.8 110300.81426.014.8481.66.661.950.0027495.00.0450.21932983298532.3491.4 112300.80426.715.1481.76.662.200.0038491.10.0530.19936343634527.2489.3 196300.81427.015.8482.56.651.920.0034491.30.0460.17226853685521.7489.5 114400.81428.338.4484.24.091.420.0044493.60.0600.47050565056536.9490.1 121450.81427.938.0484.43.281.480.0058491.10.0630.48753895389535.5487.8 212250.80429.10.4484.58.342.600.004503.90.1090.85455955596520.3491.7 118400.82427.737.9485.74.081.460.0043490.70.0460.24638623863535.3488.5 209400.80430.640.7486.14.080.800.0025494.00.0340.26544714471563.9490.4 ��NASA/TM220023The range of static wet bulb temperatures of the Icing Wedge for these ice accretion points was from 465.9 to 503.9R with a 5 particle in the flow analysis model. Note that Escortdata point 291 with a Twbs of 465.9R did not accrete due to low IWAR (0.0027), not due to Twbs (see Figure 2). The true range of Twbs for the cases with collection, but no accretion was from 487.9 to 503.9

38 R. The average difference of the Twbsfor
R. The average difference of the Twbsfor these collection, but no ice accretion points was 0.7R below the 492minimum threshold Twbs of the Icing Wedge, with a range of 16R. The static wet bulb temperature (Twbs) versus IWAR for the cases with hard ice accretion and ones which had collection, but no ice cretion, are plotted in Figure For the data point that experienced neither ice accretion nor ice collection, except a slight amount of ice at the support strut leading edge, the key parameters are listed in Table and in AppendixC ofReference 17Figure The static wet bulb temperature (Twbs) versus IWAR in the splittershroud region, for the cases with hard ice accretion and ones which had only collection, but no ice accretion (from Table Table TABLE SPLITTERSHROUD: NO ICE ACCRETION, EXCEPT ATSUPPORT STRUT (LOW Twbs Escort data p oint Altitude, K ft Flight Mach Ambient t emp., ° R Delta ISA, ° R Tt PSL Station 1 ice c loud ON, °R Pt PSL Station 1, psia Ice Water Content (IWC) IWAR Twbs splitter, shroud, fan core (6), 5 Ice mass flow into c ore, l bm/sec Kinetic energy heat f lux, Qk Conduct heat f lux, Qc, BTU/Hr Total heat f lux, Qt, BTU/Hr T wall core after 60s (6) 2.1, °R T ice particle (6) 2.1, ° R 289360.61 413.5 23.4 443.9 4.212.41 0.0061 461.5 0.050 0.147 23242324 485.3 459.6 ��NASA/TM220023Figure illustrates a data point of Table where there was no ice accretion on the splittershroud region, however some ice at the strut leading edge appears to accrete. Thestatic wet bulb temperature isR below the Icing wedge minimum threshold te

39 mperature of 492R. It is hypothesized th
mperature of 492R. It is hypothesized that even the heat flux from the splittershroud wall was not adequate to melt the particle at those operating conditions. However, a small amount of ice was noted on the strut leading edge, which may have a higher heat flux at this discrete circumferential location (shown in Figure by blue open circles). The calculated static wet bulb temperature distribution for this data point is illustrated in Figure . This data point is represented by the blue open circle in Figure and Figure Figure Ice did not collect on the splitterlip or the shroud, at data point. (Escort 289). ��NASA/TM220023Figure Ice accretion did not occur in Escort 289 which had a static wet bulb temperature 31R below the Icing Wedge minimum threshold of 492TABLE SPLITTERSHROUD: NO ICE ACCRETION (CALCULATED HOTWALL) Escort Data Point Altitude, K ft Flight Mach Ambient t emp., ° R Delta ISA , ° R Tt PSL Station 1 ce loud ON, °R Pt PSL Station 1, psia Ice Water Content (IWC) IWAR Twb s s plitter, s hroud, an core (6), 5 Ice Mass f low into c ore, l bm/sec Kinetic e nergy h eat f lux, Qk Conduct h eat f lux, Qc, BTU/Hr Total h eat f lux, Qt, BTU/Hr T wall c ore after 60s (6) 2.1, °R T ice particle (6) 2.1, ° R 15945 0.76 424.6 34.7 473.6 3.120.500.002 485.2 0.021 0.153 3581 3581 578.4 481.7 20840 0.83 427.3 37.4 485.6 4.081.00 0.0031 493.7 0.042 0.332 4829 4829 553.5 490.1 For data points that experienced no ice accretion or ice collection, the key parameters arelisted inTable . These two data points w

40 ere considered to have wall temperatures
ere considered to have wall temperatures that were too warm to support ice accretion, asshown by brown square symbols in Figure . Note that for Escort 159, there may be two reasons for no accretion: the IWAR was low, as well as the calculated Twbswas excessively high, considering the additional source of heat flux from the shroud wall. ��NASA/TM2200235K ft AltitudeThe calculated wall temperatures and static wet bulb temperatures at 5K ft altitude for data points with accretion and no accretion, versus IWAR are plotted in Figure . The calculated wall temperatures for the 5K ft points with ice accretion appear to be near, or slightly below the calculated wall temperaturesfor the high altitude points with ice accretion. The video image illustrating the ice accretion points on the splitterlip and shroud for Escort data point 243 is illustrated in Figure Figure illustrates the static wet bulb temperature distribution for Escort points 243 through the full fan, the fancore and the four stage axial compressor, as well as the particle melt ratio over a range of particle size for 3 . Ice accretion occurred at the statorshroud region at a wet bulb temperatureR below the Icing Wedge minimum threshold.Figure Calculated wall temperature and the static wet bultemperature at the splittershroud wall region at 5K ft altitude for ice accretion and no ice accretion data points (from Table Table ��NASA/TM220023Figure Splittershroud: 5K altitude ice accretionon splitterlip and shroud forEscort data point 243.Figure The static wet bulb temperature distribution and particlemelt ratio for the 5K ftaltitude Escort

41 point 243 through the full fan, the fanc
point 243 through the full fan, the fancore and the four stage axial compressor, as well as the particle melt ratio. ��NASA/TM220023For data points that experienced ice accretion at the 5K ft altitude, the key parameters are listed inTable , and are illustrated by the filled green circles in Figure Table lists the key parameters associated with the 5K ft altitude data points that did not have ice accretion, and are illustrated by the green open circles in Figure TABLE SPLITTERSHROUD: 5K ALTITUDE;ICE ACCRETION Escort ata oint Altitude, K ft Flight Mach Ambient emp., Delta ISA , °R Pt PSL Station 1, psia Ice Water Content (IWC) IWAR Twbs s plitter, s hroud, f an core (6), 5 Ice m ass f low into c ore, l bm/sec Kinetic e nergy h eat f lux, Qk Conduct h eat f lux, Qc, BTU/Hr Total h eat f lux, Qt, BTU/Hr T wall c ore after 60s (6) 2.1, ° R Tt2 Stg 2 (10), °R T ice particle (6) 2.1, °R 2500.20 432.8 68.3 12.61 5.90 0.0052 477.3 0.194 1.040 3511 3512 483.7 588.0 473.6 2450.20 434.3 66.6 12.60 12.00 0.0101 471.6 0.306 1.176 2726 2727 475.0 555.9 470.0 2460.20 434.8 66.2 12.61 6.20 0.0052 471.8 0.158 0.608 2639 2640 478.6 557.0 469.3 2430.20 440.5 60.6 12.61 10.60 0.0091 476.7 0.267 1.006 3040 3041 476.9 567.0 470.6 TABLE SPLITTERSHROUD: 5K ALTITUDE;NO ICE ACCRETION Escort ata oint Altitude, K ft Flight Mach Ambient emp., Delta ISA , °R Tt PSL Station 1 i ce c loud ON, ° R Pt PSL Station 1, psia Ice Water Content (IWC) IWAR Twbs s plitte

42 r, s hr oud, fan core (6), 5 Ice
r, s hr oud, fan core (6), 5 Ice m ass f low into c ore, bm/sec Kinetic nergy eat lux, Qk Conduct eat lux, Qc, BTU/ Total eat lux, Qt, BTU/Hr T wall c ore after 60s (6) 2.1, °R ce particle (6) 2.1, °R 2440.19 429.8 71.4 433.1 12.61 10.58 0.0089 466.5 0.281 1.161 2982 2983 469.6 463.7 2420.21 432.9 68.0 436.9 12.64 10.50 0.0092 471.1 0.274 1.023 2911 2912 474.6 468.7 ��NASA/TM220023Figure Calculated wall temperature at the splittershroud wall region versus calculated air static temperature.Figure illustrates the strong effect of the local air static temperature on the ice accretion potential in the splittershroud region. The higher air temperatures result in the collection of ice on the splitterlip, with rapid shedding a rate of several times per second, to several seconds per shed, with no accretion of hard glassy ice. However, as is evident from the data presented in the tables above there are other factors that influence ice accretion, besides the local static air temperature. Therefore, to accurately model this heat flux from the wall to the ice particle, as well as other potential heat sources, high fidelity multidisciplinary model of the entire front frame would be requiredhat includes fluid dynamic analysis as well as conjugate heat transferVariable Inlet Guide Vane (IGV) Region; Liquid Waterrom Upstream SourceAs in the splittershroud region, the Icing Wedge thresholdsare likewise not applicable in the IGV regionan indicator of accretion risk. This is likewise due to the additional heat flux through the wall at the splitt

43 ershroud region that may provide liquid
ershroud region that may provide liquid water to the IGV. During accretion on the IGV, the calculated Twbs were from 33.4to 19.4R below freezing, and the calculated static air temperatures were from 4 to 23R below freezing, and the measured metal temperature of the IGV was from 492 to 488R. Accretion on the IGV appeared to be a function of IWAR for all operating points as shown in Figure For all test points the minimum limit for iceaccretion had IWAR values above 0.008. Above this value of IWAR ice accretion occurred at low fan speeds near 7300 and 6300 RPM, and are illustrated in Figure by the black circles, and the values of key parameters are listed in Table . These were operating points where ice accretion was observed to build on the IGV without shedding. There were also operating points where the ice did not accrete, but collected briefly and rapidly shed at the IGV leading edge. These points are represented by delta symbols in Figure . For IWAR values less than 0.008, no accretion or collection occurred for any of the data points which were confirmed by video, as shown by the red square symbols in Figure ��NASA/TM220023Figure Measured IGV metal temperatures versus IWAR for all operating points with video confirmation.TABLE INLET GUIDE VANE: ICE ACCRETION; VIDEO CONFIRMATION Escort data p oint Altitude, K ft Flight Mach Ambient t emp Delta ISA, °R Tt PS L Station 1 ce loud ON, °R Pt PSL Station1, psia Ice Water Content (IWC) IWAR T wall c ore after 60s (6) 2.1, °R Twbs, IGV Station 8 Melt ratio, Station 8 ce article IGV 2.5 (8) COMDES, °R T- metal IGV 2.5 Pre c

44 e, T - m etal IGV 2.5 Post ice; 60s,
e, T - m etal IGV 2.5 Post ice; 60s, °R 284360.61412.822.9443.44.214.100.0105480.7469.60.000470.8502.1490.6 295450.77403.113.2450.53.132.950.0102481.8465.50.000465.5496.5488.0 300450.77388.31.6434.23.112.950.0102474.9458.60.000458.7497.3489.7 299450.77392.82.9439.43.122.920.0102478.4462.20.000462.3503.0491.8 156450.77403.613.7451.03.132.730.0096492.0470.20.000470.4511.2492.3 280250.60411.118.5441.06.985.800.0089476.5470.50.000470.7496.2491.0 279250.60410.119.6439.76.985.800.0089475.3470.50.000470.6494.2491.9 2420.21432.968.0436.912.6410.500.0092474.6472.60.000472.9491.6491.9 ��NASA/TM220023Figure Ice accretion on the IGV surface near the tip (Escort 284 at 36K ft altitude).Figure The static wet bulb temperature distribution and particle melt ratio for the 36K ft altitude Escort point 284 through the fancore and the four stage axial compressor, as well as the particle melt ratio. The key parameters for the Escort data points withice accretion on the IGV are listed in Table , and the complete analysis results are listed in Appendixof Reference. An example of one operating point (Escort 284) with ice accretion on the IGV is illustrated in Figure Figure illustrates the static wet bulb temperature distribution for Escort points 284 through thefancore and the four stage axial compressor, as well as the particle melt ratio over a range of particle size for 3 . Ice accretion occurredat the IGV near the tip at a wet bulb temperature that was 22.4below the Icing Wedge minimum threshold (492R). ��NASA/TM220023Ice collection and rapid shedding was observed on the IGV during two test data points

45 (123, 153) at measured metal temperatur
(123, 153) at measured metal temperatures that were well above freezing, and are illustrated with black delta symbols in Figure , and summarized in Table . While the value of IWAR for these two data points was above 0.008, the engine the local values of static air temperature were significantly higher than the air temperatures for the points with ice accretion. The higher static air temperatures at the IGV for these two data points are caused by higher inlet total air temperatures at Station 1 as well as significantly higher fan speeds. At these two operating conditions the ice collected on the IGV leading edge and rapidly shed,as shown inFigure Figure illustrates the static wet bulb temperature distribution for Escort points 123 through thefancore and the four stage axial compressor, as well as the particle melt ratio over a range of particle size for 3 . Ice collection and rapid shedding occurred at the IGV leading edge near the tip at a wet bulb temperature that was within the Icing Wedge minimum threshold. No ice accretion occurred.Through video confirmation it was observed that with IWAR values less than 0.008, no ice accretion occurred on the IGV, only ice collection at the 5K ftaltitude point. These operating points are shown by the red square symbols in Figure and the values of the key parameters are listed in Tabl, as well as in AppendixC of Reference 17TABLE INLET GUIDE VANE: ICE COLLECTION WITH RAPID SHEDDING; VIDEO CONFIRMATION Escort data p oint Altitude, K ft Flight Mach Ambient t emp Delta ISA, °R Tt PSL Station 1 ce loud ON, °R Pt PSL Station1, psia IWAR T wall core after 60s (6) 2.1

46 , °R Twbs, IGV Station 8 Melt ratio
, °R Twbs, IGV Station 8 Melt ratio, Station 8 ce article IGV 2.5 (8) COMDES, °R T-m etal IGV 2.5 Pre i ce, T- metal IGV 2.5 Post ice; 60s, °R 153450.81399.49.5452.23.280.0099495.5475.50.000475.8522.6498.0 123400.81414.524.6469.34.080.0089509.9490.70.159491.7556.6514.0 Figure Ice collection on the IGV leading edge near the tip with rapid shedding (Escort 123 at 40 K ft altitude). ��NASA/TM220023Figure The static wet bulb temperature distribution andparticle melt ratio for the 40K ft altitude Escort point 123.TABLE INLET GUIDE VANE: NO ICE COLLECTIONNO ICE ACCRETION; VIDEO CONFIRMATION Escort data p oint Altitude, K ft Flight Mach Ambient t emp Delta ISA, Tt PSL Station 1 Ice c loud ON, °R Pt PSL Station 1, psia Ice Water C ontent (IWC) IWAR T wall core after 60s (6) 2.1, °R Twbs, IGV Station 8 Melt ratio, Station 8 ce article IGV 2.5 (8) COMDES, °R T- metal IGV 2.5 Pre ce, T- Metal IGV 2.5 Post ice; 60s, °R 293450.77402.812.9450.53.122.280.0081490.7468.10.000468.3511.3493.1 100250.40436.710.4450.95.862.930.0061506.3490.20.121491.2538.5512.1 149400.61406.917.0437.73.491.540.0049499.2466.80.000466.8510.4495.8 103250.65434.75.0471.77.272.500.0042519.7496.10.508491.7543.7511.1 147400.62397.87.9428.53.491.280.0041501.3463.70.000463.8507.2496.7 157360.61422.332.4453.64.181.460.0039511.6475.30.000475.4515.4496.3 105250.65434.54.9471.87.282.360.0037522.5491.40.110491.7529.5500.8 108350.80424.831.7479.95.251.520.0036533.3495.60.555491.7560.8523.5 159450.76424.634.7473.63.120.500.002578.4486.00.000486.8560.2549.2 Figure illustrates the computed ice part

47 icle temperature and measured IGV metal
icle temperature and measured IGV metal temperatures,60 secafter the ice cloud was turned on, versus IWAR. The black and blue symbols represent ice accretion or collection as confirmed by video. The red symbols represent data points without ice accretion, as confirmed by video. Note that the calculated particle temperatures do not include the enthalpy rise that was transferred to the particle from the heat flux through the wall, near the splittershroud region upstream of the IGV. Therefore the calculated particle temperatures iFigure to Figure are significantly below the measured IGV metal temperature, which is near freezing for most data points where ice accreted on the IGV. ��NASA/TM220023Figure Computed ice particle temperature and measured IGV metal temperatures versus IWAR. The black and blue symbols represent ice accretion or collection. The red symbols represent points without ice accretionFigure Measured IGV metal temperatures pre and post ice cloud on versus IWAR. The black and blue symbols represent ice accretion or collection. The red symbols represent data points without ice accretion.Figure illustrates the measured IGV metal temperatures pre and post ice cloud on versus IWAR. In most cases, the measured IGV metal temperature reduced significantly after the ice cloud was turned on. The black and blue symbols represent ice accretion or collection as confirmed by video. The red symbols represent data points without ice accretion, as confirmed by video. ��NASA/TM220023Figure illustrates the computed static air temperature at the IGV and the measured IGV metal temperatures, versus IW

48 AR. The black and blue symbols represent
AR. The black and blue symbols representice accretion or collection as confirmed by video. The red symbols represent data points without ice accretion, as confirmed by video.Figure illustrates the computed static air temperature and the computed ice particle temperature versus IWAR. The black and blue symbols represent ice accretion or collection as confirmed by video. The red symbols represent data points without ice accretion, as confirmed by video.Figure Computed static air temperature and measured IGV metal temperatures versus IWAR. The black and blue symbols represent ice accretion or collection. The red symbols represent data points without ice accretion.Figure Computed static air temperature and ice particle temperatures versus IWAR after ice cloud on. The black and blue symbols represent ice accretion or collection. The red symbols are data points without ice accretion. ��NASA/TM220023The remaining data points that were analyzed had no video available to confirm whether there was any ice accretion or collection on the IGV. The key calculated and measured parameters for these data points are listed in Table , and the complete flow analysis results are listed in AppendixC of ReferenceTABLE INLET GUIDE VANE: NOVIDEO AVAILABLE FOR CONFIRMATION Escort Altitude, K ft Flight Mach Ambient t emp, Delta ISA, °R Tt PSL Station 1 t ce loud ON, °R Pt PSL Station 1, psia Ice Water Content (IWC) IWAR T wall core after 60s (6) 2.1, Twbs, IGV Station 8 Melt ratio, Station 8 ice p article IGV 2.5 (8) COMDES, °R T- metal IGV 2.5 Pre Ice, °R T- metal IGV 2.5 post ice; 60s, °R 2

49 89360.61413.523.4443.94.212.410.0061485.
89360.61413.523.4443.94.212.410.0061485.3462.90.000462.8487.5484.3 282360.61412.222.7442.64.211.000.0026507.9466.90.000466.8500.2493.2 287360.6412.222.3443.14.222.480.0066493.1473.10.000473.3516.8498.6 283360.61412.722.8443.34.212.000.0051492.4468.10.000468.1501.1491.0 298450.78396.26.4444.23.142.950.0103482.5465.30.000465.5507.7492.7 291450.77403.613.7451.43.120.750.0027519.2466.90.000466.9511.5498.0 292450.77403.113.2451.53.121.330.0047503.3467.70.000467.8512.1494.5 297450.76407.217.3454.93.122.930.0105488.7471.60.000471.8515.3493.2 208400.83427.337.4485.64.081.000.0031553.5494.60.556491.7574.1535.1 276250.41436.87.3451.36.122.930.0059505.4490.30.134491.3539.4511.2 187250.41437.57.9451.96.122.930.0059507.1490.80.159491.7539.6510.0 193300.61429.318.2461.15.562.890.0065508.2491.60.242491.7548.6515.5 122300.60429.918.4461.25.572.880.0065509.5491.90.260491.7548.2510.5 191300.61429.918.7461.75.572.900.0066508.6491.80.256491.7548.4517.0 188300.6431.119.8462.75.572.840.0064511.5492.90.318491.7549.2509.3 214250.81412.4 16.9 466.08.344.000.006506.4495.50.442491.7548.7509.0 213250.8418.3 11.3 472.38.352.600.004517.6498.20.615491.7555.1525.4 124350.81420.327.3475.25.252.890.007515.0495.90.543491.7560.8521.9 110300.81426.014.8481.66.661.950.0027532.3495.80.552491.7546.2508.2 112300.80426.715.1481.76.662.200.0038527.2492.40.190491.7531.6499.3 196300.81427.015.8482.56.651.920.0034521.7492.50.202491.7532.4499.8 114400.81428.338.4484.24.091.420.0044536.9494.70.552491.7571.0529.8 121450.8427.938.0484.43.281.480.0058535.5492.50.387491.7574.7529.9 212250.80429.10.4484.58.342.600.004520.3504.81.000501.2567.7534.2 118400.82427.737.94

50 85.74.081.460.0043535.3492.10.260491.755
85.74.081.460.0043535.3492.10.260491.7555.1512.0 209400.80430.640.7486.14.080.800.0025563.9494.80.579491.7574.5539.7 2440.19429.8 71.4 433.112.61 10.58 0.0089469.6467.80.000468.2487.7489.3 2500.20432.8 68.3 436.212.615.900.0052483.7478.70.000479.2504.7495.9 2450.20434.3 66.6 437.912.60 12.00 0.0101475.0473.10.000473.5492.6491.8 2460.20434.8 66.2 438.412.616.200.0052478.6472.90.000473.2492.5491.3 2430.20440.5 60.6 444.012.61 10.60 0.0091476.9478.30.000478.6498.6493.3 ��NASA/TM220023There was no direct measurement of the metal temperatures near the splittershroud region, however the calculated metal temperatures there were compared to the measured metal temperatures at the IGV and are shown in Figure . Included on the plot is a reference line (black dashdot line) depicting a line with a slope of 1 for the two temperatures, as well as a linear least squares curve fit of the calculated splitterhroud metal temperatures (blue dashed line). The trend of the calculated wall metal temperature closely tracks the measured IGV metal temperature. This indicates that the calibrated heat transfer model of the splittershroud region is adequate for comparison of calculated wall temperatures at the various test points. HPC Stage 1 Stator Region; Liquid Water from Upstream SourceFor certain operating pointsin the test matrix, ice was expected to accrete on HPC stage 1 stator based on the pretest analyses. However, during testing it appeared that ice did not accrete there, since the measured metal temperatures of the variable stator were 61.6 to 79.8 R above freezing, as shown inTable . This was possibly

51 due to additional heat conducted to the
due to additional heat conducted to the particles from the upstream splitterlip and shroud wall. In addition the energy imparted by HPC rotor 1 further increased the air temperature, thus raising thetemperature of the particle entering stator 1. Note that there were no video cameras in this region.Figure Comparisonof the IGV measured metal temperature to the calculated metal temperature at the splittershroud region. The trend of the calculated metal temperature closely tracks the measured IGV metal temperature.TABLE HPC STAGE 1; STATOR:NO VIDEO AVAILABLE FOR CONFIRMATION Escort ata oint Altitude, K ft Flight Mach Ambient emp Delta ISA, °R Tt PSL Station 1 ce loud ON, R Pt PSL Station 1, psia Ice Water Content (IWC) IWAR Twbs, HPC tator 1 Stat. 13 T-m etal HPC Stator 1, re ce, T- Metal IGV 2.5 Post Ice; 60s, °R 153450.8199.49.5452.23.282.900.0101509.1631.1553.6 149400.62437.717.0406.83.491.540.0049501.8623.0566.8 157360.61453.632.4422.14.181.460.0040506.4614.3571.8 ��NASA/TM220023Figure Percentage difference between the CD model and the Escort test data as a function of Escort number.UncertaintyKey HURE engine performance parameters from the CD system model were compared to the Escort measurement data in order to validate the model. Figure illustrates the percentage difference of the parameters between the measured data and the CDas a function of Escort data number, which increased chronologically. Variances of up to 6percentwere noted for certain engine performanceparameters. The analysis conducted in this study is based on the CD estimate of the flow through the core. These di

52 fferences may affect the analysis result
fferences may affect the analysis results with COMDESMELT, thus adding to the uncertainty.The current meanline flow analytical capabilityfor calculating the values of the key icing parameters lacks the radial distribution at the blade row edgeshusits accuracy for estimating the span wise location of accretionalong the fanstator needs to be improved, or higher fidelity modeling needs to be implementedFuture WorkA higher fidelity computational capability (twodimensional streamline curvature flow analysis, or full threedimensional CFD)is required in order to calculate the values of the key icing parameters through the span of a blade/stator at multiple radial locations. Additionally, the heat flux through the nonadiabatic wallscan be a major factor in ice accretion. The differences between the CD model and the measured Escort data need to be reconciledThis would also reduce theuncertainty of the COMDESMELT model, which relies on input from both. The measurement uncertainty can be decomposed into the sampling uncertainty and the systematic uncertainty. For example, the sampling uncertainty can be due to variations in altitude, inlet temperature, fan speed and flight Mach number between test data points. The systematic uncertainty can be caused by possible measurement errors in the instrumentationand data acquisition. All the easurement uncertainties can propagatewhich cause variation in the accuracyof the model ��NASA/TM220023Summary/ConclusionsThe Honeywell Uncertified Research Engine (HURE) has been tested in the Propulsion System Laboratory (PSL) at NASA Glenn Research Center with ice crystal cloud ingesti

53 on over a range of simulated altitudes,
on over a range of simulated altitudes, ambient temperatures, and engine operating conditions (varying flight Mach number and fan speed). A computational process utilizing the meanline (COMDESMELT) aerodynamic compressor flow analysis code along with the Honeywell provided customer deck were utilized for the posttest analysis of the enginetest data. The HURE engine testing indicates that ice accretion occurred differently in the fanstator, than it does near the splitterlip and shroud region, as well as on the compressor inlet guide vane. Ice accretion on the nonmetallic fanstator vanes occurs within the range of wet bulb temperature thresholds of the Icing Wedge (492 to R) where it was expected to occur, and did not have any icing when the Twbs was outside these Icing Wedge thresholds. The accretion here appears to be an adiabatic process. In the fanstator region the static wet bulb temperature thresholds of the Icing Wedge, proved to have a 94percentaccuracy as an indicator of icing risk. Although the meanline flow analytical capability lacks the radial distribution of particles, and key icing parameters, it can be an effective tool for estimating their bulk values. Early during the testing phase of the HURE, realtime analysis of the data indicated that the splitterlip and shroud region of the gooseneck may have been heated, however the heat source was not well understood. The PSL test plan was adjusted accordingly for subsequent test data points. In order to induce ice accretion near the splittershroud region, the target value of static wet bulb temperature was reduced byR to a value near 468R. This proved to be a su

54 ccessful technique for forcing ice to ac
ccessful technique for forcing ice to accrete in that region, in spite of the suspected nonadiabatic walls. During posttest analysis, a simple heat transfer model was developed in order to calculate the wall temperature in the splittershroud region at all operating conditions. The heat transfer model was incorporated into the compressor flow analysis code, for an ordermagnitude calculation of wall temperature. In this nonadiabatic region, the Icing Wedge wet bulb temperature thresholds were not applicable as an indicator of icing risk due to the additional heat flux through the walls. The ice accretion on the variable inlet guide vanes (IGV) of the compressor appeared to be a strong function of the IWAR (icewater flow rate to air flow rate ratio). Ice accretion on the IGV occurred at values of IWAR above 0.008, whereas no ice accretion on the IGV occurred below that value. In the IGV region, the Icing Wedge thresholds were not applicable as an indicator of icing risk due to the source of liquid water from upstream (splittershroud). Nonadiabatic compressor flow path walls require a fully coupled multidisciplinary analysis of the conjugate heat transfer through the walls, the air flow, as wellas the thermodynamic state of the ice particle, in order to determine the static wet bulb temperature distribution in the flow field. NASA/TM—2018-220023 43 References Mason, J. G., Chow, P., Fuleki, D. M., “Understanding Ice Crystal Accretion and Shedding Phenomenon in Jet Engines Using a Rig Test,” GT2010-22550. Mason, J. G., Grzych, M., “The Challenges Identifying Weather Associated With Jet Engine Ice Crystal Icing,” SAE 2011-38