Document
2019-01-1920
Scaling Evaluation of Ice-Crystal Icing on a Modern Turbofan Engine in PSL Using
the COMDES-MELT Code
Jen-Ching Tsao
Ohio Aerospace Institute turbofan engine in order to develop numerical simulation models of
Abstract
various fidelity, with the goal to determine possible engine icing onset conditions and locations; predict ice accretion locations, This paper presents preliminary ice-crystal icing (ICI) altitude scaling profiles and sizes; and evaluate/address design challenges to mitigate evaluation results of a Honeywell Uncertified Research Engine the risks [1]. The NASA Engine Icing Research team has continued (HURE) that was tested in the NASA Glenn Research Center developing its understanding of ice-crystal icing through a number of Propulsion Systems Laboratory (PSL) during January of 2018. This full scale engine test campaigns conducted in the PSL [2-7] and a engine geometry features a hidden core design to keep the core less series of fundamental ICI physics experimental studies conducted on exposed. The engine was fitted with internal video cameras to a NACA 0012 airfoil in the NRC Research Altitude Test Facility observe various ice buildup processes at multiple selected locations (RATFac) and the NASA Propulsion Systems Laboratory (PSL) [8- within the engine core flow path covering the fan stator, the splitter- 13]. A simulated altitude engine icing test facility like the PSL lip/shroud/strut, and the high pressure compressor (HPC) variable enables highly controlled environmental simulations at relevant flight inlet guide vane (IGV) regions. The potential ice accretion risk was conditions to be examined. Instrumentation development to acquire pre-determined to occu r by using NASA’s in -house 1D Engine Icing details on the cloud generation in the facility and the ability to fully Risk assessment code, COMDES-MELT. The code was successful in instrument an engine allows researchers better understand of the predicting the risk of ice accretion in adiabatic regions like the fan- underlying physical mechanisms in order to further improve and stator of the HURE at specific engine operating points. However at validate the existing in-house engine ice accretion simulation and several operating points during the test, liquid water was observed icing risk prediction codes. Those early research works formed the running along the shroud toward the variable IGV of the HPC regions basis of the icing risk criteria used in the development of the NASA with an air temperature well below freezing, thus no particle melting in-house 1D Icing Risk tool [14-18].
could have occurred due to heating from the air alone. It was reasoned that other sources of heat were present in that region. To NASA, in collaboration with Honeywell and with the support of the account for these heat sources the inlet total temperature was adjusted International Ice Crystal Consortium, had successfully conducted two to give a wet bulb temperature of 24 F below the standard minimum full engine test campaigns on an unmodified ALF502R-5 engine wet bulb temperature of 492 R to allow ice to accrete in the splitter- configuration [2-7]. Those studies gave an initial understanding of lip/shroud/strut region, which was determined from a reference case possible ICI features in a relevant engine icing environment.
where hard ice was observed in that region. With that adjustment the However this engine was an older design and featured a heated COMDES-MELT code was successful in providing operating points spinner which provided an additional source of liquid. This current where there was a risk of ice accretion during the test campaign. In study examines the Honeywell Uncertified Research Engine (HURE) addition to calculating possible conditions at different selected lower which is a different turbofan engine design with a hidden core. This altitudes, simulations were run to determine potential inlet conditions engine was never in production so t here isn’t any known IC I issues.
that could lead to ice-crystal accretion along the prescribed stations Therefore this study served as a blind test that provided the NASA where the cameras were available. From there, scaled test conditions Engine Icing Research team a chance to assess how effectively the were determined by best matching the following three icing related 1D Icing Risk tool could predict icing risk and how the proposed parameters of the reference condition: (1) the local air total wet bulb altitude scaling method could be used for full engine ICI testing at or temperature, (2) the local ice crystal cloud melt ratio and (3) the near sea-level altitudes. This could help develop a method to test at engine fan face ice/water to air mass flux ratio of the ice crystal sea-level facilities since availability and cost are important factors in cloud. Instantaneous images taken from the time-lapsed movies of ice an altitude simulation facility consideration.
buildup were used along with the relevant thermodynamic data of air, water vapor and local icing condition to help evaluate how closely The altitude scaling consideration for ICI is a condition scaling the proposed altitude scaling method could be used in ground based method used to determine the necessary scaled test conditions that test facility to duplicate selected reference ICI features observed at enables similar ice accretion features from the reference condition specific location inside this engine at different scale altitudes.
chosen at selected location inside the engine compression system.
Discussions on observed limitation for engine icing scaling Recently a thermodynamic model was developed by Tsao et al. [4] application from this test campaign and needed improvement are specifically for studying engine ICI phenomena and the model provided. A scaling test procedure to help identify potential ICI risk analysis has been examined with early fundamental ice crystal icing conditions and possible ice accretion locations of a new turbofan physics experimental data [8-9] and with selected data from the two engine is evaluated in PSL.
previous engine ICI tests [1-2]. The ALF502 engine known to have ice-crystal accretion in the EGV2 TE region had helped calibrate the
Introduction
1D Icing Risk tool to find the icing conditions at lower altitudes and illustrate that altitude scaling is possible in PSL to simulate the ice A key goal of the NASA Engine Icing Research is to better crystal accretion features. It may be of interest to note that in order to understand the complex physics and possible interaction of the ice- observe accretion at low altitude, the air static wet bulb temperature crystal icing (ICI) phenomena that take place inside a modern T wb, ∞ had to be decreased for the ALF502 while T wb, ∞ was increased Page 1 of 9 for the HURE test. This difference between the two engines could be profile, i.e. fixed T and MR , the freezing fraction n decreases as wb,0 0 due to different heat transfer effects between high altitude and 5K ft. (IWC /ρ ) increases. This non-dimensional term ( IWAR ≡ IWC /ρ ) i a i a The focus of this study is to, using the 1D Icing Risk tool, show suggests that for engine ICI problem the air density effect is also preliminary evaluation results of the proposed altitude scaling method important. For the same IWAR value ingested by the engine, the local at possible scaled icing conditions and accretion locations for the IWC i value increases as altitude goes from 45K ft to 5K ft due to heavily instrumented HURE engine ICI test conducted in PSL in increasing air density value. It was necessary to increase the IWC i to 2018. maintain the IWAR when going from 45K ft to 5K ft, see Table 5 below. This effect was first noticed and reported by Veres et al . [17] in their recent engine icing analysis work in terms of the ice-water
Proposed Scaling Parameters for Engine ICI
flow rate to air flow rate ratio, IWAR , from the heavily instrumented ALF502 (S/N LF11) engine icing test data. This suggests that the Key Equations for Similarity Parameter Consideration most important parameter affecting the ice crystal accretion process is the freezing fraction n .
This aforementioned thermodynamic model for engine ICI study was primarily based on previous scaling work by Anderson [19] and (2) Melting Dominated Regime Anderson and Tsao [20] using the Messinger’s steady -state surface energy balance analysis [21] for super-cooled liquid water icing in The expression of the melting fraction of ice particle at stagnation both Appendix C and Super-cooled Large Drop (SLD) regimes. The point region can be written in the following form expressions for key surface energy balance equations and the associated similarity parameters involved will be presented here MR MR without much discussion. Therefore, readers who are interested in the * = ( ) T T m c (2) , 0 0 0 wb s physical descriptions and detailed derivations of these parameters are VWC VWC * gain gain referred to [4, 7, 19 and 20] and the references given therein. Several (1 ( )) (1 ( )) MR MR b IWC IWC key modifications were made to the surface energy balance equations i i in order to account for possible interactions between the local air flow within the low/high pressure compressor flow passage and the Similar to the freezing dominated process, Eq. (2) also shows that for ingested ice crystals from the engine inlet.
melting dominated process on icing surfaces the melting fraction m is the most important parameter. Additionally, it is noticed that the (1) Freezing Dominated Regime most effective variables from the test facility to control the resulting ICI feature are the IWAR and T .
wb,0 The final expression of the freezing fraction of liquid melt at stagnation point region can be written in the following form
Experiment Description
* Facility Description
( ) T T c n (1)
0 ,0 0 s wb * b The HURE engine ice-crystal icing test was conducted in the NASA Glenn Research Center Propulsion Systems Laboratory Cell 3 (PSL- in which T denotes the total air wet bulb temperature and is given wb,0 3). PSL is a direct-connect altitude test chamber that was modified as with an icing cloud generation system. Details on the facility capabilities can be found in [22] and [23] and are briefly highlighted p p in Table 1.
P
, , v ws v w
RH
T T P V h
0 G PSL can generate ICI conditions through a spray nozzle freeze out
, T T
,0 wb v
p methodology which is described in [24]. The facility icing operation
1 2 P c h
, v ws
0 , p air c
envelope is illustrated in Figure 1. Prior to the test, a cloud calibration
0.622 T T
0 was conducted to characterize the cloud for the requested icing conditions and gain further facility experience by exploring the PSL parameter space. Details of this calibration can be found in a recent and publication by Van Zante et al. [24].
( ) m MR
IWC
imp f Table 1. PSL-3 Capability * i a , b
0
h IWC VWC
c i gain a
2
RT M
IWC
*
i a . c
0
( ) MR IWC VWC
f i gain a
* Evidence that probe under-measured From Eq. (1), it is clear that both parameters T and MR can affect # wb,0 Particles larger than ~ 60 microns are NOT fully glaciated the accretion process however they are not unique. There are many likely combinations of T and MR that could lead to the same wb,0 freezing fraction n . It is also noted that for a given local thermal Page 2 of 9 needed information on the particle MVD size of ice crystals ingested into the core.
Scaling Evaluation of Ice-Crystal Icing for a New
Engine
The HURE engine is a prototype engine for research and development purpose with many notional design features that has not entered for production. Therefore it does not have any known ICI information or flight data of field events that could help the HURE engine test team to quickly identify possible range of atmospheric and engine operating conditions as well as the most likely IC icing locations (e.g. highest IC cloud impinging mass flux locations) inside the engine compression system. This provides an opportunity for the team to develop a test procedure in PSL to identify potential engine ICI test points in Appendix D conditions and then obtain corresponding altitude engine performance and icing effect data for analytical modeling tool development.
Figure 1. PSL-3 Icing Envelope For determining the ICI test matrix the NASA team utilized its in- house 1D engine icing risk assessment code, COMDES-MELT, to Test Article Description generate atmospheric and engine operating conditions at different altitudes that could enable ice-crystal icing to occur at several The Honeywell Uncertified Research Engine (HURE) is a turbofan selected stations within the compression system, as highlighted in engine that features a hidden core design. This design with an inlet Figure 3. Details of this extensive computational prediction analysis gooseneck feature keeps the core less exposed, see Figure 2. Details work was reported in 2017 by Veres et al . [27] prior to the HURE about the test experimental set-up is described in [25].
test entry. After the test, Veres et al. also performed a post-test data analysis of 57 Escort test points that were taken at distinct operating conditions in PSL at altitudes from 5K ft to 45K ft. The detailed analysis results for these test data points with the Honeywell Customer Deck and with the COMDES-MELT codes were reported Vane Metal TCs in Reference [28]. 4 IGV S trut LE Pt/Tt 3 Loc (4 Struts) Figure 2. HURE Engine Cross-Section (Courtesy Honeywell Engines) Instrumentation Description Figure 3. Selected ICI Meridional Stations within HURE Compression System (Ref [27]) The engine was fully instrumented with typical aero-research instrumentation. This paper will describe the instrumentation relevant As shown from Eqns. (1) and (2), for possible similarity of ice crystal to the local IC icing feature assessment for scaling evaluation accretion on specific unheated surface location within the engine purpose. Static pressure data were obtained on the front frame hub compression system flow passage, it is necessary for the scale test to leading and trailing edge and along the HPC outer shroud at rotors 2 simulate the geometry, the flow field, the ice/water particle and 3. Total temperature data were acquired at the engine inlet, at trajectories, the total water catch, the heat transfer and, probably, the four circumferential positions at the splitter-lip strut leading edge at surface phenomena of the desired reference icing condition.
three radial locations and aft of the fan in the bypass duct with two However most of the needed information in order to conduct proper T2 sensors. Metal temperature data were acquired at four scaling calculation for this HURE engine were not known or circumferential locations at the HPC IGV, and stator vanes 1 and 2.
available at the time. An alternative approach to find potential scale Humidity measurements were obtained at the core inlet and in the test conditions was established as follows: HPC exit. A simple layout of temperature measurement instrumentation used in this study is highlighted in Figure 2.
(1) A potential IC icing risk condition at 45K ft was selected as the reference condition for this scaling Important measurements of ice crystal particle size distribution after study. The reference IC accretion features will be passing through an engine fan were obtained and the details regarding evaluated at those prescribed camera view locations the method and analysis is given in [26]. This data would provide Page 3 of 9 inside this engine. It covers the fan stator region at station 3-5, the splitter-lip strut LE at station 6 and the HPC IGV LE at station 8 in COMDES-MELT ’s station number notation shown in Figure 3.
(2) Three scaled test conditions with similar engine ICI risk at altitudes of 36K, 25K and 5K ft with best matching engine fan-core and LPC/HPC operation characteristics (ex: the engine fan face Mach number Figure 5. Escort 156, Screenshots of Fan Stator (Stator 2 & Stator 4 View) M ff , the corrected engine fan face mass flow rate W cor , Showing No Ice Accretion Formed on the EGV.
and the corrected engine fan rotational speed N1 ) cor from the reference condition were calculated using the COMDES-MELT code to best match the following key icing parameters of the reference condition at those TE LE three selected locations: a. the local total wet bulb temperature T wb,0 , IGV b. the local IC cloud melt ratio MR , and Figure 6. Escort 156, Screenshots of Splitter - Lip Shroud (left) and IGV c. the engine fan face IWAR of the IC cloud (as Surfaces (right) Showing Firm Ice Accretion and Runback Water.
an approximation of the local IWAR value).
When the PSL were able to estimate the Table 2. Two Potential Reference ICI Conditions at 45K ft.
TWC value ingested into the core with the Light Extinction Probe (LEP) system [25], it should match the local IWAR . Esc Alt M T T P IWAR TW C MVD FLT amb PL PL - 3 3 Rdg K ft P sia (10 ) g/m F F m (3) A parametric sweep of IWC, MVD, N1 and T pl around each proposed scale condition will be performed M M M M M C C C (within the facility operation limit) to ensure sufficient coverage of local reference icing conditions at prescribed icing risk locations would be simulated to 121 45.1 0.81 -31.6 24.4 3.27 5.5 1.5 28 allow reasonable evaluation of the proposed scaling method.
156 45.2 0.77 -56.3 -9.5 3.13 9.7 2.8 29 The Reference Condition ICI Features During the first two days of HURE ICI test campaign, the team tested The Escort is the PSL data recording system. The Escort reading (Esc a number of potential ICI test points, chosen from Reference [27], in rdg) 121 is a case where ice-crystal melting is dominant inside the 1-3 minutes short spray in the PSL to identify icing cases to be used engine core flow passage. Quick ice buildup with infrequent ice as ICI anchor points for detailed parametric studies.
shedding was observed on the pressure side of the fan stators (EGV) as shown in Figure 4, further into the splitter-lip shroud and strut LE For the reference condition at 45K ft, two potential ICI conditions region rapid ice shedding with discrete small white ice deposits with were considered, Escort readings 121 and 156. The corresponding no discernable growth was observed along the splitter lip, and no ice atmospheric and engine operating conditions run in PSL are listed in was observed in the HPC IGV area where the surface metal Table 2 which includes facility measured parameters (M) like temperature was well above freezing.
altitude, flight Mach number, ambient air temperature, plenum air temperature and pressure and some facility calculated parameters (C) The Escort reading 156, however, is a case where IC melting is not as like IC cloud total water content, IC cloud MVD size and IWAR . The strong as case 121 and freezing of IC melt on the surface is occurring team evaluated the corresponding ICI accretion features and the inside the engine core flow passage. No ice accretion was observed extent of icing coverage over those three camera view areas and then on the fan stator vanes where the IC particle temperature was still picked the best one for scaling study. sub-freezing with very little melt and the fan stator vanes were also dry and mildly cold. The ambient temperature was 25 F colder than Escort reading 121 with almost twice the TWC resulting in a higher IWAR . Thus no ice was collected or accreted on the fan stators as shown in Figure 5. The COMDES-MELT analysis also predicted that there would be no ice accretion in the fan stator region for this case Fan 156. In the splitter-lip and strut LE region very glazy ice accretion started to form on the splitter-lip region and lots of water runback EGV was seen on the splitter-lip shroud surface area, see the left image in Figure 6. Also very little amount of firm ice accretion was formed on the IGV LE tip area as well as on the pressure side TE region of the vanes as illustrated in the right image of Figure 6.
Figure 4. Escort 121, Screenshots of Fan Stator (Stator 2 & Stator 4 Views) Showing Quick Ice Buildup on the Pressure Side of EGV.
Page 4 of 9 Table 3. The T , T , T Values and the Observed ICI Features at Station 3, It was noted in Reference [28] that ice accretion was observed on the 0 wb0 IGV 5, 6 & 8 for Those Two Potential Reference ICI Conditions at 45K ft.
front frame components near the splitter-lip at lower static wet bulb temperatures than was expected, based on the Icing Wedge minimum Esc T T T T MR T T Metal , threshold of 492 R. In addition, the accretion occurred at static air 0,3 wb0,5 0,6 wb0, 6 6 IGV IGV R dg F F F F Metal , Spray on, @ temperatures well below freezing, thus no or very little particle Pre Spray, 60 sec , melting could have occurred due to heating from the air alone. The F F aluminum front frame may have received heat from additional M COMDES M COMDES COMDES M M 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 121 115.5 52.0 117.3 46.4 0.17 115.0 70.1 adiabatic process which was assumed in the 1D simulations. During testing, in order to compensate for the lack of a heat transfer model in 156 35.2 21.8 49.8 19.5 0.0 51.5 32.6 the COMDES-MELT code, the target static wet bulb temperature Icing ( T ) for ice to accrete in the front frame region was reduced to 468 wb ,∞ Fan Stator Splitter - Lip Shroud HPC Area R. This was 24 R below the Icing Wedge minimum threshold of (EGV) & Strut LE IGV 492 R. The new target T wb ,∞ was determined by the analysis of one of the operating points where ice accreted at the splitter-lip and 121 X shroud region. Using COMDES-MELT, new testing conditions were rapidly derived prior to further testing, and the test matrix was 156 X modified. This was successful in enabling ice to accrete in the front frame components (splitter-lip and shroud region). For post-test data analysis, a simple bulk heat transfer model was developed to estimate the wall metal surface temperature [28]. This was done in order to Table 4. The Matching Similarity Parameters between Reference and Scale Test compare it to previous engine tests which had measured wall Conditions temperatures between 492 R to 501 R during ice accretion.
Esc T MR IWAR However the observation of water runback on the splitter-lip surface wb0, 6 6 - 3 Rdg F (10 ) area was different from what the COMDES-MELT calculation of Escort Rdg 156 case has shown that the IC particles would not experience any melting from the airflow in the core flow passage 156 19.5 0 9.7 from the splitter-lip to HPC IGV regions (i.e. MR = 0 from stations 5- 8). As mentioned earlier there was possible unknown heat transfer mechanism not accounted for by the COMDES-MELT code in this 284 19.1 0 11.0 area that could promote melting of IC particles. Jorgenson et al. [28] has provided an order of magnitude estimate of possible heat transfer model hypothesis in this region (i.e. assuming a non-adiabatic wall) 279 21.7 0 9.0 to promote the observed IC melting as seen in Figure 6. It was decided then for finding all scaled test conditions the COMDES- 242 24.2 0 9.1 MELT code was adjusted by the selected reference condition.
Table 3 shows the facility measured air total temperatures at stations 3 & 6 and the pre-spray metal surface temperature at station 8 that collectively provide the initial compression system thermal energy The Scaled Condition ICI Features for possible melting of ingested IC cloud, and also includes the COMDES-MELT calculated total wet bulb temperatures at station 5 & 6 and the measured metal surface temperature after a minute of IC Three scaled test conditions at altitudes of 36K, 25K and 5K ft were spray at station 8 which represent the final thermal equilibrium states best matching the splitter- lip and strut LE station’s total wet bulb of the ingested IC cloud and the compression system. Clearly the temperature (i.e. T wb0,6 ), IC particle melt ratio (i.e. MR 6 ) and the thermal data from the Escort Rdg121 case exhibits a melting engine fan face IWAR of the reference condition ( Esc Rdg 156 case).
dominant ICI features but the Escort Rdg 156 case, with 25 F lower inlet temperature, displays a potential low freezing ICI features. Table 5. The Reference and Scaled Test Conditions at 45K, 36K, 25K and 5K ft.
It is worth noting that in finding the best ICI case as the reference condition, it was recognized from the videos that there are several Esc Alt M FLT T amb T PL P PL IWAR TWC MVD - 3 3 key areas for ICI in the fan stator (adiabatic region), splitter-lip and Rdg Kft F F Psia (10 ) g/m m shroud region (non-adiabatic region) and the HPC IGV (possibly affected by melting ice from non-adiabatic region) as highlighted in M M M M M C C C Figures 4 and 6. David Rigby at el. [29] presents a numerical study of IC ingestion into HURE with more details given in the paper that 156 45.2 0.77 -56.3 -9.5 3.13 9.7 2.8 29 corroborates these observations. Based on the reference ICI features 284 36.2 0.61 -46.9 -16.2 4.21 11.0 4.2 N/A described in Table 3 and further considering the air mass flow rate constraint found at running 5K ft altitude simulation in PSL requires 279 25.0 0.60 -49.6 -19.7 6.99 9.0 5.8 55 that the fan speed and flight Mach number must be reduced significantly in order to duplicate the reference ice accretion, 242 4.92 0.20 -26.2 -22.9 12.60 9.1 10.7 52 therefore the Escort Rdg 156 case is chosen to be the reference condition for scaling study of ICI.
Page 5 of 9 The best matching similarity parameters for ICI scaling application Finally the Table 6 shows, for the reference and all three scaled test were shown in Table 4 and the resulting atmospheric and engine conditions, the measured air total temperatures at stations 3 & 6, the operating conditions run in PSL were shown in Table 5. Melt ratio measured pre-spray metal surface temperature at station 8, the was not considered a key parameter for icing risk for this region, calculated total wet bulb temperatures at station 5 & 6 and the since the melting of the ice particle not coming from the air but was measured metal surface temperature after one minute of IC spray at most likely coming from their impact with a heated surface. station 8.
Figure 10. Escort 243 (left) and 245 (right) cases, Screenshots of Splitter - Lip Figure 7. Escort 284, Screenshots of Splitter - Lip Shroud (left) and IGV Shroud Surfaces Showing Firm Glaz e Ice Accretion with Some Running (right) Surfaces Showing Firm Ice Accretion and Some Running Wet.
Wet.
Similar to the reference case ICI features, those three scaled test conditions all exhibit freezing dominant ICI features in general.
However for the near sea level 5K ft altitude test point (i.e. Esc Rdg 242), the IC cloud melting before reaching to the splitter-lip region is not as much as the reference condition of the other two higher altitude scaled test points have. By raising the ambient air temperature 5 F warmer or increasing the TWC by 15% seems to provide the needed portion of IC melt in the splitter-lip surface area.
Figure 8. Escort 279, Screenshots of Splitter - Lip Shroud (left) and IGV (right) Surfaces Showing Firm Ice Accretion and Some R unning Wet.
From the preliminary evaluation it is found that the PSL has developed sufficient control of the air temperature, the TWC and MVD of an IC spray, as well as the cloud uniformity for most of the atmospheric conditions within its icing envelope (as shown in Figure 1) for various ICI simulations. Acceptable scaling results were obtained even at its 5K ft operation limit point. The scaling evaluation of ICI features on this engine in PSL suggests that altitude scaling is possible if the engine core flow passage (mainly in the low pressure compression section of the engine) surface heat transfer feature is available and properly accounted for by the N ASA’s in - Figure 9. Escort 242, Screenshots of Splitter - Lip Shroud (left) and IGV house 1D Engine Icing Risk assessment code, COMDES-MELT This (right) Surfaces Showing Firm Ice Accretion with Very Little Running Wet.
would enable PSL or a sea level engine test facility to simulate the important ice crystal accretion features and the associated ICI effects Similar to the aforementioned reference condition ICI features the on engine performance. In particular, the scaling test procedure used Escort Rdg 284, 279 and 242 are cases where ice-crystal freezing is in PSL for this engine test has helped identify potential icing risk dominant inside the engine core flow passage as well. No ice deposit conditions and possible ice accretion locations for a new turbofan or accretion was observed on the fan stator vanes (due to low ambient engine design where no ICI information is available in advance.
air temperature). In the splitter-lip and strut LE region firm glaze ice accretion started to form on the splitter-lip region and some runback It should be noted that the PSL engine icing facility continues water was observed on the splitter-lip shroud surface area for Escort working on improving its ability to have better icing cloud quality.
Rdg 284 & 279, see Figures 7-8. But for the Escort Rdg 242 case it The in-house higher fidelity numerical icing simulation codes are showed very little wetting on the splitter-lip region where only small under development and they are still limited by current measurement firm glaze ice accretion was observed in Figure 9. The calculated wet capability and larger test data uncertainty for rigorous model bulb temperatures were below freezing for these three cases. Again validation purpose. In addition, there is no robust measurement this is a region where other heated components in the flow path may method to acquire data in the engine core flow passage locations contribute greatly to the melting of ice particles when they impact where initial ice buildup is first occurred due to the complexity and these heated surfaces.
harsh engine environment. With these constraints in mind, the images and the time-lapsed movies of ice formation from the cameras As stated in the proposed scaling test procedure a parametric sweep represent the only available visual characterization of resulting ice of two most ICI sensitive facility variables (i.e. TWC and T pl ) were shapes. The qualitative ice shape comparison allows one to assess performed around the case 242 to see how much change was needed how well this altitude scaling method could replicate the reference ice for each variable in order to reproduce the selected reference ICI crystal accretion features observed.
features at this altitude of 5K ft.
In the Escort Rdg 243 case the ambient temperature was raised by 5 F and in Escort Rdg 245 case the TWC was increased by 15%. Both conditions resulted in firm glaze ice accretion observed with some surface water runback on the splitter-lip region, see Figure 10.
Page 6 of 9 Table 6. The T , T , T Values and the Observed ICI Features at Station 3, 0 wb0 IGV 24 R temperature drop in air static wet bulb 5, 6 & 8 for Reference and All Three Scaled Test Conditions temperature T in the splitter-lip shroud strut region, wb ,∞ and Esc T T T T T T 0,3 wb0,5 0,6 wb0,6 MR 6 IGV IGV iii. the engine fan face IWAR of the IC cloud Rdg F F F F Metal, Metal, 3. A parametric sweep of IWC, MVD, N1 and T pl around each Pre Spray on, proposed scale condition was performed (within the facility Spray , @ 60 sec , operation limit) to ensure sufficient coverage of local reference F F icing conditions at prescribed icing risk locations.
M COMDES M COMDES COMDES M M The scaling evaluation of ICI features on this engine in PSL 156 35.2 21.8 49.8 19.5 0.0 51.5 32.6 illustrates that altitude scaling is possible for PSL or a sea level 284 35.8 20.6 41.8 19.1 0.0 42.4 31.2 engine test facility to simulate the important ice crystal accretion features and the associated ICI effects on engine performance. It was 279 36.4 21.8 36.3 21.7 0.0 34,5 32.2 also shown from 5K ft scaled test results that due to the additional heat from the splitter-lip region, the scaled test condition (i.e. Escort 242 35.8 26.9 33.2 24.2 0.0 32.0 32.2 Rdg 242 case) had to make some change, either by raising the Icing Fan Stator Splitter - Lip Shroud HPC ambient air temperature 5 F warmer (i.e. Escort Rdg 243 case) or Area (EGV) & Strut LE IGV increasing the TWC by 15% (i.e. Escort Rdg 245 case) to replicate the firm glaze ice accretion with some surface runback water on the splitter-lip region that was observed in the reference condition (i.e.
156 X Escort Rdg 156 case). Thus for future engine ICI studies, it is strongly suggested that the dry-air performance and heat transfer 284 X characteristics of the low pressure compression system should be available for complete IC icing risk analysis. Finally, a scaling test 279 X procedure was proposed and evaluated in PSL to help identify potential icing risk conditions and possible ice accretion locations for a new turbofan engine design where no ICI information is available 242 X in advance.
References
Conclusions
1. Flegel, A. B., “ Ice Crystal Icing Research at NASA Glenn Preliminary evaluation results of the proposed altitude scaling Research Center," 9th AIAA Atmospheric and Space method for studying a full scale turbofan engine ICI features is Environments Conference, AIAA 2017-4085, June 2017.
presented from a heavily instrumented engine test conducted in the 2. Oliver, M. J. "Validation Ice Crystal Icing Engine Test in the NASA Glenn Research Center Propulsion Systems Laboratory (PSL) Propulsion Systems Laboratory at NASA Glenn Research during January of 2018.
Center," 6th AIAA Atmospheric and Space Environments Conference, AIAA-2014-2898, June 2014.
It is shown from the evaluation that there exist two distinct types of 3. Goodwin, R.V., Dischinger, D.G., “Turbofan Ice Crystal Rollback Investigation and Preparations Leading to Inaugural ice formation for this engine: (1) when the icing surface is dominated Ice Crystal Engine Test at NASA PSL- 3 Facility,” 6th AIAA by the freezing of IC melt from fully or partially melted impinging ice crystals, the ice structure is formed from accretion with strong Atmospheric and Space Environments Conference, AIAA-2014- 2895, June 2014.
adhesion to the surface, for example the Escort reading 156 case, and 4. Tsao, J., Struk, P. M., and Oliver, M. J., "Possible Mechanisms (2) when the icing surface is dominated by the further melting of ice crystals, the ice structure is formed from accumulation of un-melted for Turbofan Engine Ice Crystal Icing at High Altitude," 6th AIAA Atmospheric and Space Environments Conference, ice crystals with relatively weak bonding to the surface, for example AIAA-2014-3044, June 2014.
the Escort reading 121 case.
5. Flegel, A.B., Oliver, M.J., “Preliminary Results from a Heavily Instrumented Engine Ice Crystal Icing Test in a Ground Based A proposed scaling test procedure was evaluated for this engine ICI Altitude Test Facility, ” 8th AIAA Atmospheric and Space study: Environments Conference, AIAA 2016-3894. NASA/TM-2016- 29132.
1. A potential IC icing risk condition at 45k ft was selected as the 6. Goodwin, R.V., Fuleki, D., “Turbofan Ice Crystal Rollback reference condition for this scaling study. The reference IC Investigation and Preparations Leading to the Second, Heavily accretion features were evaluated at those prescribed camera Instrumented, Ice Crystal Engine Test at NASA PSL-3 test view locations inside this engine.
Facility,” 8th AIAA Atmospheric and Space Environments 2. Three scaled test conditions with similar engine ICI risk at Conference, AIAA-2016-3892, June 2016.
altitudes of 36k, 25k and 5k ft were calculated using the 7. Tsao, J., “ Preliminary Evaluation of Altitude Scaling for COMDES-MELT code that was adjusted by the selected Turbofan Engine Ice Crystal Icing,” 9th AIAA Atmospheric and reference condition to best match the following three key icing Space Environments Conference, AIAA 2017-4086, June 2017.
parameters of the reference condition at the HPC IGV location: 8. Struk, P., Currie, T., Wright, W. B., Knezevici, D. C., Fuleki, i. the local total wet bulb temperature T , wb,0 D., Broeren, A., Vargas, M., and Tsao, J. "Fundamental Ice ii. the local IC cloud melt ratio MR , however noticed that Crystal Accretion Physics Studies," SAE 2011 International for this engine due to the additional heat from the Conference on Aircraft and Engine Icing and Ground Deicing, splitter-lip region non-zero MR of IC cloud from SAE Technical Paper 2011-38-0018 or NASA/TM-2012- COMDES-MELT calculation was not possible for all 217429, June 2011.
cases since inlet conditions were adjusted to get a Page 7 of 9 9. Currie, T. C., Struk, P. M., Tsao, J., Fuleki, D., and Knezevici, Atmospheric and Space Environment Conference, AIAA-2018- D. C. "Fundamental Study of Mixed-Phase Icing with 3969, June 2018.
Application to Ice Crystal Accretion in Aircraft Jet Engines," 25. Flegel, A., Agui, J., King, M., Tsao, J. C. and Chen, R. C., “Ice 4th AIAA Atmospheric and Space Environments Conference, Crystal Effects on a Hidden Core Turbofan Engine in an AIAA-2012-3035, June 2012. Altitude Simulation icing Facility,” Oral Presentation, 2019 10. Struk, P. M., Bencic, T., Tsao, J., Fuleki, D., and Knezevici, D. SAE International Conference on Icing of Aircraft, Engines, and C. "Preparation for Scaling Studies of Ice-Crystal Icing at the Structures, 19ICE-0137, Minneapolis, MN, June 17-21, 2019.
NRC Research Altitude Test Facility," 5th AIAA Atmospheric 26. King, M., Flegel, A., Bencic, T., Manin, J., Baumgardner, D., and Space Environments Conference, AIAA-2013-2675 and Dischinger, D., Wienkes, L. and Wilson, B. “Particle Size NASA/TM-2013-216571, 2013. Measurements from the 2018 Honeywell Uncertified Research 11. Struk, P. M., Bartkus, T. P., Tsao, J. C., Currie, T., and Fuleki, Engine Test in the NASA Propulsion System Laboratory,” 2019 D. "Ice Accretion Measurements on an Airfoil and Wedge in SAE International Conference on Icing of Aircraft, Engines, and Mixed-Phase Conditions," SAE 2015 International Conference Structures, 19ICE-0147, Minneapolis, MN, June 17-21, 2019.
on Icing of Aircraft, Engines, and Structures, SAE Technical 27. Veres, J. P., Jorgenson, P. C. E., “Prediction and Analysis of Ice Paper 2015-01-2116, June 2015. Accretion in a Research Turbofan Engine With Ice Crystal 12. Struk, P. M., Tsao, J. C., and Bartkus, T. P., “Plans and Cloud Ingestion at Simulated Altitudes,” NASA/TM -2017- Preliminary Results of Fundamental Studies of Ice Crystal Icing 219724 (availability restricted to U.S. government and NASA Physics in the NASA Propulsion Systems Laboratory,” 8th contractors) AIAA Atmospheric and Space Environments Conference, 28. Jorgenson, P. C. E., Veres, J. P., Bommireddy, S. R., Nili, S., AIAA-2016-3738, June 2016. “Analysis of the Honeywell Uncertified Research Engine 13. Struk, P. M., Ratvasky, T. P., Bencic, T. J., Van Zante, J. F., (HURE) with Ice Crystal Cloud Ingestion at Simulated Altitudes King, M. C., Tsao, J. C., and Bartkus, T. P., “An Initial Study of (Publi c Version),” NASA/TM— 2018-220023.
the Fundamentals of Ice Crystal Icing Physics in the NASA 29. Rigby, D. L., Wright, W. B., Flegel, A. and King, M., Propulsion Systems Laboratory,” 9th AIAA Atmospheric and “ Simulation of Ice Particle Breakup and Ingestion into the Space Environments Conference, AIAA-2017-4242, June 2017. Honeywell Uncertified Research Engine (HURE),” 2019 SAE 14. Veres, J. P., Jorgenson, P. C. E., “Modeling Commercial International Conference on Icing of Aircraft, Engines, and Turbofan Engine Icing Risk with Ice Crystal Ingestion,” 5th Structures, 19ICE-0159, Minneapolis, MN, June 17-21, 2019.
AIAA Atmospheric and Space Environments Conference, AIAA- 2013-2679 and NASA/TM-2013-218097.
Contact Information
15. Jorgenson, P. C. E., Veres, J. P., Coennen, R., “Modeling of Commercial Turbofan Engine with Ice Crystal Ingestion; Jen-Ching (Paul) Tsao, Principal Research Scientist Follow- On,” 6th AIAA Atmospheric and Space Environments Work phone: (216) 433-2411 Conference, AIAA-2014-2899 and NASA/TM-2014-218496.
E-mail: jenching.tsao-1@nasa.gov 16. Veres, J.P., Jones, S.M., Jorg enson, P.C.E., “Performance Affiliation: Ohio Aerospace Institute Modeling of Honeywell Turbofan Engine Tested with Ice Crystal Ingestion in the NASA Propulsion System Laboratory,” SAE 2015 International Conference on Aircraft and Engine
Acknowledgments
Icing and Ground Deicing, SAE Technical Paper 2015-01-2133.
17. Veres, J.P., Jorgenson P. C. E., Jones, S. M., “Modeling of This work was supported under the NASA Advanced Air Vehicles Highly Instrumented Honeywell Turbofan Engine Tested with Program, Advanced Air Transport Technology Project and the Ice Crystal Ingestion in the NASA Propulsion System Aeronautics and Evaluation Test Capability Project. The first author Laboratory,” 8th AIAA Atmospheric and Space Environments was supported under a NASA ARTS contract. The author wishes to Conference, AIAA-2016-3895, June 2016.
acknowledge the financial support for this work. The author also 18. Veres, J.P., Jor genson, P.C.E., Jones, S.M., Nili, S., “Modeling wishes to acknowledge the special help and guidance received from of a Turbofan Engine with Ice Crystal Ingestion in the NASA Mr. Joe Veres and Dr. Phil Jorgenson for providing COMDES- Propulsion Systems Laboratory,” ASME IGTI 2017 Turbo Expo , MELT analysis results and the contributions from the rest of the Charlotte, NC, June 26-30, 2017, GT2017-63202.
Engine Icing Research team for this work. The author would like to 19. Anderson, D. N., “Manual of Scaling Methods,” NASA /CR– thank the entire PSL staff for their dedication and support of this test.
2004-212875, March 2004.
The author also wishes to thank the Honeywell team for their 20. Anderson, D. N. and Tsao, J. C., “Ice Shape Scaling for Aircraft contributions during the engine test.
in SLD Conditions,” NASA/CR -2008-215302, DOT/FAA/AR- 07/55, September 2008.
Definitions/Abbreviations/Subscripts
21. Messinger, B.L., “Equilibrium Temperature of an Unheated Icing Surface as a Function of Airspeed,” J. Aer -on. Sci., vol. 20 * no. 1, January 1953, pp 29 – 42. m odified relative heat factor b 22. Griffin, T. A., Lizanich, P., and Dicki, D. J., "PSL Icing Facility Upgrade Overview," 6th AIAA Atmospheric and Space * c s pecific ratio of ice particle Environments Conference, AIAA-2014-2896, 2014.
kinetic heating to latent heat 23. Dicki, D., Thomas, Q., Oliver, M., Kowalewski, J., Poljak, P., absorbed from melting Zimmerle, K., Rosine, B., Rachow, P., Shrewsbury, C., “ Propulsion Systems Laboratory Customer Guide Update,” 2019 EGV e xit guide vane SAE International Conference on Icing of Aircraft, Engines, and Structures, 19ICE-0160, Minneapolis, MN, June 17-21, 2019.
24. Van Zante, J., Ratvasky, T., Bencic, T., Challis, C. and Timko, h c onve ctive heat - transfer c E., “Update on the NASA Glenn Propulsion Systems Lab Icing coefficient [ cal/ s m K ] th and Ice Crystal Cloud Characterization – 2017,” 10 AIAA Page 8 of 9 h g as - phase mass - transfer T i cing s urface temperature [ °C ] G s coefficient [ g / s m ] T wb wet bulb temperature [[ °C ] IGV inlet guide vane a ir static temperature [ °C ] T IWAR normalized IWC = IWC/ρ a total water content [g/m ] TWC IWC ice water content [g/m ] V air velocity [kt] LWC liquid water content of melt [g/ m ] VWC vapor water content gained = gain MMD ice crystal median mass diameter 3 VWC – VWC [ g/ m ] t i [μm] W fan face mass flow rate [ lbm/s ] MR ice crystal melt ratio, = LWC /IWC t i latent heat of fusion [ cal/g ] f MVD water drop median volumetric diameter [μm] latent heat of evaporation [ cal/g ] v local Mach number M ∞ air density [kg/m ] a m mass flux of ice/liquid - water imp particle impinged per unit time, Subscripts: lbm/ft s cor corrected m melting fraction at stagnation ff fan face region i inlet, initial or cloud off N1 fan speed [rpm] pl in the plenum core speed [rpm] N2 t target, terminal or cloud on n 0 freezing fraction at stagnation region s at the surface p air static pressure [N/m ] 0 stagnation or total value saturation water vapor pressure in p v,w 2 local static or ambient condition ambient [N/m ] p s aturation water vapor pressure at v,ws icing interface [ N/m ] RH r elative humidity SH specific humidity [g/kg] Page 9 of 9