1 Status Update Mohan Gupta, Ph.D. Asst. Chief Scientist ASCENT FALL MEETING October 13-15, 2015 Seattle, WA
1
Status Update
Mohan Gupta, Ph.D.Asst. Chief Scientist
ASCENT FALL MEETINGOctober 13-15, 2015
Seattle, WA
2
Current Two-Phase ASTM Fuel Approval Process
• OEMs have identified key Figures Of Merit (FOM) to determine acceptable combustion performance.
• Altitude Relight• Lean Blowout• Cold Start
• Tier 3/4 testing is critical for evaluating FOMs. Testing costs increase significantly as fuels transition from Tier 1/2 to Tier 3/4 testing performed by the OEMs
• Fuels approved to date have chemical compositions similar to petroleum based jet fuel• HEFA, FT and DSHC (at 10% blend) fuels performed as expected.
• But DSHC at 20% pushed composition beyond typical range and exhibited unacceptable performance and was not approved.
• New generation of candidate fuels have very different chemical composition and will demand extensive testing and resources
3
Overview of NJFCP Program
NJFCP is relating fuel properties to combustion FOM.
Program uniqueness:• Integrated systemwide
approach involving all stages of testing and modeling areas for identical conditions
• Real-time communication and share of info among all 6 areas (experimentalists and modelers) and OEMs
• Brings state of the art knowledge, computer capabilities, and engineering experience together
Fit-for-purpose testing
Area 7: Program interface and integration
ASTM Tier 3/4
ASTM Tier 1/2
Vision: Develop an experimental and analytical capability to facilitate OEM’s evaluation of fuel physical and chemical properties on engine operability and to streamline ASTM fuels approval process.
4
Improved OEM Screening of Fuels with NJFCP Integration
Acceptable Combustor Operability?
Yes
Redesign/Reengineer Fuel Development
Pathway
No
Scope of Tier 3/4 Testing Determined by
NJFCP Results
NJFCP: Initial Fuel Screening at a representative operating
condition using OEM designed 1-Cup Rig linking fuel to the FOM
NJFCP: Detailed Fuel Testing & Combustion Modeling at an extended range of conditions using
OEM designed 1-Cup Rig to study FOM
Expa
nd &
Em
ploy
Gai
ned
Kno
wle
dge
Bas
e
Benefits: Early fuel screening, targeted Tier 3 and 4 tests, and increased OEM confidence
Fuel Usage Avoided: 20K galsTest Costs Avoided: $4MTime Saved: 3 yrsOverhead Costs Avoided: ????
Fuel Usage Avoided: 6K galsTest Costs Avoided: $2.3MTime Saved: 1 yrOverhead Costs Avoided: ????
5
OverviewThe NJFCP program has made great progress in the past 5 monthsGetting the experiments online quicklyShowing some early fuel sensitivities Working to demonstrate repeatability and evaluating uncertainties
Main PointsMaking great progress! Need to move more quickly to cold and sub-atmospheric conditionsMore clarity needed on modeling planIf funding becomes an issue may want to consider a shift of resources to testing Program should be open to adding fuelsSooner we get to extreme conditions the better (i.e. cold air and fuel, low pressure)Need to determine how blends of the C fuels affects the results
SuccessIf the OEM team, over the course of the program, gains insight and broadens an understanding of fuel effects on combustion (as per OEMs, this is already happening on a daily basis in their combustion operability related activities)
Exceeds ExpectationsIf the program team develops tools/models that show the ability to simulate fuel effects and trends seen in the chosen experiments
OEM Review and FeedbackFrom Mid-Year Review Meeting
6
AprMar Oct
Program TimelineEarly Fuel Screening, (Testing), Year-End Demo of Fuel Effects (Modeling)
Oct 27-30, 2014 MACCCR Meeting
Dec 3-4, 2014Kickoff Meeting
Fuels Survey, Screening & downselection
May end, 2015 Semiannual Meeting: Progress towards meeting success criteria and adaptations needed going forward
NJFCP: Year 2 Starts
Jan Feb May Jun Jul Aug Sep Nov Dec Jan
Mid-month telecon
Month-end SC telecon
Feb Mar
Oct 19-21, 2015 MACCCR Meeting
Fuel Screening Tests
Fuel-dependent Model Development
Detailed data acquisition
Model validation Fuel Effects
Verification
FAA Review meetingDecision Point: Program continuity in Year 2
CAAFI and ACENT meeting
TESTINGMODELING
CRC meeting
Oct 22-23, 2015: NJFCP Year-end meeting, 2015Status towards meeting success criteria
7
Program Sponsors, Contributors, Performers and Industry Members
National Jet Fuels Combustion ProgramCRC Aviation MeetingMay 6, 2015A total of 34 entities4 additional universities under AFOSR program also contribute to the program.
STEERING COMMITTEE(Federal, OEMs, University PIs)
Guidance
Fed Gov’t‐FAA‐AFRL‐AFOSR‐NASA‐DLA‐Navy‐DOE‐ARL‐NIST
‐Funding‐Scientific FoundationTest Facilities‐Fuels
Industry‐Honeywell‐GE‐Pratt & Whitney‐Williams‐Rolls‐Royce‐Fuel Producers‐Parker Hannifin
‐Chem/Kinetics Modeling‐Engine Operability‐Fuel Evaluation Methodology‐Reduced cost
NJFCPASCENT Universities: GaTech, UDRI, UIUC, Stanford, Purdue, OSU Non‐ASCENT: Princeton,
UConn, USC
Other Contributors:NASA, AFRL, NIST, ARL, NRC
Canada, DLR, OEM, Sandia Lab, LLNL, University of Sheffield
ASCENT Advisory Committee Members(CAAFI, Boeing, Shell, Gevo)
GuidanceInternational: NRC, DLR, Univ. Sheffield
Information Exchange
8
Budget Details
Funding Area FAA AFRL$ DLA Energy NavyOEMs 500 500NJFCP Testing (ASCENT) 1,150 1,300 250
NJFCP Modeling (ASCENT) 750 ------NJFCP Integration (ASCENT) 100Other Testing & Contract Support ------ 171 500 200Sub Total 2,500 1,971 750 200
Grand Total 5,421
Agency ContributionsFAA (Testing areas 1, 3 and 5) 1,270
AFRL (Testing area 6 + misc.) 1,650
NASA (All modeling areas: 2, 4 and 5) 1,103
DLA (misc.) 500
Navy (misc) 200
Grand Total 4,723
Year 1 Funding ($K)
Year 2 Resources ($K) Additional Contributions• AFOSR (consistently investing in fundamental combustion
areas through its core and STTR. programs: $1.4M/year in efforts directly related to NJFCP areas on leveraged bases, not included in tables left)
• ARL (in-house activities, $550K)• DOE (in-house activities at National Labs and possible support
to secure fuels for testing)• NASA (in-house activities)• NIST (in-house activities) • NRC (in-house activities, $500K)• DLR (In-house activities funded by EU ECLIF program)• Univ. Sheffield (in-house activities)
$AFRL spends additional funds (that are not included here) to procure fuels and rig development and maintenance for the program.
Besides FAA, many other federal agencies and international partners are significant contributors to the program.
FAA CENTER OF EXCELLENCE FOR ALTERNATIVE JET FUELS & ENVIRONMENT
Project manager: Mohan Gupta, FAA
Joshua Heyne, University of DaytonMeredith Colket, Contractor
National Jet Fuels Combustion Program (NJFCP)
Projects 25-30, 34
October 13-15, 2015Seattle, WA
Opinions, findings, conclusions and recommendations expressed in this material are those of the author(s)and do not necessarily reflect the views of ASCENT sponsor organizations.
10
ASCENT Project PIs and Key Contributors
• Area 1: Ron Hanson (Stanford), Tom Bowman (Stanford), Dave Davidson (Stanford), Shock Tube and Flow Reactor Studies.
• Area 2: Hai Wang (Stanford), Chemical Kinetics Model Development and Evaluation.
• Area 3: Tim Lieuwen (Georgia Tech), Jerry Sietzman (Georgia Tech), Wenting Sun (Georgia Tech), David Blunck (Oregon State), Fred Dryer (Princeton), Tonghun Lee (Illinois Urbana-Champaign), Advanced Combustion.
• Area 4: Suresh Menon (Georgia Tech), Matthias Ihme (Stanford), Tiangfeng Lu (UConn), Alejandro Briones (Dayton), Wenting Sun (Georgia Tech), Combustion Model Development and Evaluation.
• Area 5: Robert Lucht (Purdue), Paul E. Sojka (Purdue), Scott Meyer (Purdue), Carson Slabaugh (Purdue), Jay Gore (Purdue), Atomization Tests and Models.
• Area 6: Scott Stouffer (Dayton), Steven Zabarnick (Dayton), Tonghun Lee (Illinois Urbana-Champaign), Referee Combustor.
• Area 7: Josh Heyne (Dayton), Med Colket (contractor), Coordination.
11
Take-Aways/Key Points
• NJFCP program is targeted to improve and streamline current ASTM fuel approval process
• Year 1 Accomplishment Summary:
• Experimentally screened fuels, demonstrated fuel effects, and performed detailed measurements for model comparisons
• Modeling teams have simulated detailed fundamental experiments and multi-dimensional multi-phase experiments.
• OEMs are fully involved and guiding the program direction
• Community-wide national and international participation
• Leveraging interagency and international support
• Year 2 – Developing additional testing capabilities and iterating on modeling methodology.
12
Mapping of NJFCP Testing to FOMs
Yr 1 Primary Focus- Testing: Lean Blow-off (LBO), cold start, validation dataModel: Capability demonstration for LBO
Yr 2 Primary Focus- Testing: LBO, relight and validation data - old and new geometriesModel: Semi-Quantitative LBO simulations, demo for ignition
Figures of Merit (FOM)
Area Experiments Data/ObservationsLBO at
Approach
Cold and Ground Start
HighAltitude Relight
1Shock Tube • Ignition delay times, pyrolytic and oxidation product yields (limited) X X X
Flow Reactor• Detailed product distribution and ratios under pyrolysis and oxidation
(fuel rich) X
3
Liquid fuel spray rig • Phi at lean blow‐off; flame structure (shape/dynamics) data XPrevaporized ignition rig • Probability of (spark) ignition as a function of phi1 XSpray ignition rig • Probability of (spark) ignition as a function of phi X X
Turbulent flame speed• Turbulent flame speeds as function of phi from sub‐ to super atmospheric
conditions X X X
5 Atmospheric spray rig• Drop diameters, mass distributions of spray, at ambient plus w/ cooled
fuel and airX X
Pressurized spray rig• Drop diameters, mass distributions of spray, at elevated pressures and air
preheatX
6 Referee rig (OEM design)
• Phi at lean blow‐off; flame structure (shape/dynamics) data, probability of spark ignition approaching cold start conditions as function of pressure
X X
1Phi = Equivalence ratio = fuel/air ratio divided by ideal ratio
Year 1Year 2Both Year 1&2
13
Fuel Candidates and Screening• Reference Fuels Required to Characterize Rig and Engine Fuel
Response• Category A: Three Conventional (Petroleum) Fuels
--“Best” case (A-1) --“Average” (A-2) --“Worst” case (A-3)• Category C: Six “Test Fluids” With Unusual Properties
• C-1: low cetane, narrow boiling (downselected)• C-2: bimodal boiling, aromatic front end• C-3: high viscosity• C-4: low cetane, wide boiling• C-5: narrow boiling, full fuel (downselected)• C-6: high cycloparaffins (OEMs also prefer)
140
160
180
200
220
240
260
280
300
0 20 40 60 80 100
Tem
pera
ture
, C
D86 % Distilled
"flat"
bimodal
"low cetane bimodal"
"low cetane wide boiling"
"high viscosity"
"high cycloparaffins"
A3: low H/C, high viscosity, high flash (within experience base)
Boiling range plot
C-1 and C-5 were selected for detailed study in Year 1. C-6 was removed from consideration due to availability.
14
Mapping Figures of Merit (FOM)
Typical Engine Operating Space
T3-P3 Regimes of Interest
Lean Blow Out 2-4 atm, 400-450 KCold Start ~ 1 atm, 255-350 KHigh Altitude ≥ ¼ atm, ≥ 230 KIgnition
Target ConditionsP3, T3
Primary FOM:• Lean Blow Out• Cold Start• High Altitude
Ignition
T3 and P3 are the combustor inlet conditions typical for gas turbine engines.
FAA CENTER OF EXCELLENCE FOR ALTERNATIVE JET FUELS & ENVIRONMENT
Project manager: Mohan Gupta, FAALead investigator: R. K. Hanson, Stanford University
Co-Investigators: C. T. BowmanStaff/Students: D. F. Davidson, S. Banerjee, Y. Y. Zhu, T. C. Paris
Shock Tube and Flow Reactor Studiesof the Kinetics of Jet Fuels
Project 25Area 1
October 13-15, 2015Seattle, WA
Opinions, findings, conclusions and recommendations expressed in this material are those of the author(s)and do not necessarily reflect the views of ASCENT sponsor organizations.
16
Area 1: Chemical Kinetic ExperimentsRole Within NJFCP
• Provide shock tube/laser absorption and flow reactor experiments for a fundamental kinetics database for jet fuels. Experiments are designed to reveal the sensitivity of combustion properties to fuel composition for the ultimate use in simplifying the alternative fuel certification process.
• Shock tube and flow reactor measurement results are used directly by Prof. Hai Wang (Area #2) in the development validation and refinement of the HyChemmodel for jet fuels.
17
Area 1: Chemical Kinetic ExperimentsResults
• Accomplishments– Using shock tube/laser absorption methods we acquired fuel,
ethylene, and methane time-histories for all 9 FAA fuels during pyrolysis.
– In flow reactor pyrolysis experiments this set of species was extended using gas chromatography measurements to include, in particular, C3, C4 and aromatic species.
– We extensively examined shock-tube ignition delay times of 9 different fuels over a wide range of temperatures (700-1200 K) in an effort to provide the FAA with sufficient information to allow down-selection to a smaller test set.
• Program Takeaways– Shock tube and flow reactor measurements uniquely provided
critically-needed constraints on the HyChem model
18
Shock Tube Ignition Delay Time and Speciation Measurements
Shock tube/laser absorption speciation measurements provide early time (10 s to 2 ms) kinetic constraints that were directly applicable to the development of the HyChem model. Shock tube ignition delay times provided kinetic targets for the validation of the HyChemmodels (Area #2) currently being developed for the FAA fuels. The largest sensitivity to fuel composition was seen in the low temperature NTC (negative temperature coefficient) regime.
0.00.20.40.6
0.81.0
CH4
0.80% A1/Argon1202 K, 1.55 atm
Frac
tion
[%] C2H4
Product Evolution Ignition Delay TimesA‐3
A‐1
10
1
0.1
Igni
tion
Del
ay T
ime
(ms)
0.8 1.0 1.2 1.41000/T (1/K)
0 500 1000 1500 20000.0
0.5
1.0
1.5
2.0
C2H4 CH4i-C4H8 M
ole
Time [s]
0.92% C1/Argon1176 K, 1.40 atm
C3H6
Products and Ignition Delay Show Differences Between Petroleum and Cat. C Fuels
C1 – low cetane, narrow boilingC2 – bimodal boilingC3 – high viscosity
19
Comparative Flow Reactor Pyrolysis Yields
0
5
10
15
20
25
30
35
40
CH₄ C₂H₄ C₃H₆ i-C₄H₈ C₄H₆ C₂H₆ C₂H₂
Category AGevo(C1)Virent(C6)Bimodal(C2)POSF 12341POSF 12344POSF 12345
(C3)(C4)(C5)
Mol
e Pe
rcen
t of F
uel C
arbo
n@ 2
0 m
s
90 percent of the carbon atoms originally in all of the fuels investigated is inseven small hydrocarbons.
max
min
Flow Reactor Pyrolysis Yields Show Differences as Well
POSF 11498 (C1)POSF 10279 (C6)POSF 12223 (C2)
CH4 C2H6 C3H6 i-C4H8 C4H6 C2H6 C2H2
C1 – low cetane, narrow boilingC6 – cycloalkaneC2 – bimodal boilingC3 – high viscosityC4 – low cetane, wide boilingC5 – narrow boiling, full fuel
FAA CENTER OF EXCELLENCE FOR ALTERNATIVE JET FUELS & ENVIRONMENT
Project manager: Mohan Gupta, FAALead investigator: Hai Wang, Stanford University
Staff/Students: R. Xue, D.-P. Chen
Hybrid Approach to Chemical Kinetics Model Development and Evaluation
Project 26Area 2
October 13-15, 2015Seattle, WA
Opinions, findings, conclusions and recommendations expressed in this material are those of the author(s)and do not necessarily reflect the views of ASCENT sponsor organizations.
21
Area 2: Kinetic Model DevelopmentRole Within NJFCP
• To develop a reduced-order reaction model to capture most important combustion properties of three Cat-A reference jet fuels, including pyrolysis intermediate distributions, ignition delay, flame extinction and flame speed.
• To develop reduced-order reaction models for down-selected Cat C fuels (C1 & C5) and to demonstrate the similarity and differences between the Cat A and C fuels.
• Coordinate with Area 1 to obtain shock tube and flow reactor data for model development
• Interface with Area 4 in model reduction for CFD simulations.
22
Area 2: Kinetic Models Results
• Accomplishments– Developed reaction models for three Cat A fuels, and two Cat C
fuels (C1 & C5)– Demonstrate the applicability of the hybrid approach to
combustion chemistry modeling of complex, multi-component real fuels
• Program Takeaways– As far as heat release and induction-zone chemistry is
concerned, the three Cat A fuels behave almost identically in their behaviors
– The combustion properties of the C1 fuel can be notably different from those of the Cat A fuels. Compared to the A fuels and under comparable conditions, the C1 fuel undergoes faster pyrolysis but produces intermediates that are more resistant to oxidation.
23
Selected Results – the A2 model
• Pyrolysis model was developed on the basis of shock-tube and flow-reactor data;• The combined model (fuel pyrolysis and foundational fuel chemistry) was tested
against experimental ignition delay and flame speeds.
Product Evolution Ignition Delay Laminar Flame Speed
0.00
0.01
0.02
0.03
0 500 1000 1500
Mol
e Fr
actio
n
Time, t (s)
C2H4
CH4
Run 250.750% POSF10325 (A2) in ArT5 = 1329 K, p5 = 12.72 atm
102
103
104
0.6 0.7 0.8 0.9
POSF10325-4%O2-Ar
Igni
tion
Del
ay, s
)
1000K/T
10
20
30
40
50
60
70
80
0.20 0.25 0.30 0.35 0.40
A generic JP-8
0.6 0.8 1.0 1.2 1.4
Lam
inar
Fla
me
Spee
d, S
u (c
m/s
)
Fuel-to-Oxygen Mass Ratio, mfuel /mO2
o
POSF10325 (A2)T0 = 403 K, p = 1 atm
Equivalence Ratio,
Detailed Kinetic Model Replicates Combustion Characteristics Well
24
Selected Results – the A1 & A3 fuels
A1
A3
Other Petroleum Fuels Match A-2 Model Well
0.00
0.01
0.02
0 500 1000 1500
Mol
e Fr
actio
n
Time, t (s)
C2H4
CH4
Run 50.690% POSF10264 (A1) in ArT5 = 1322 K, p5 = 13.20 atm
0.00
0.01
0.02
0 500 1000 1500
Mol
e Fr
actio
n
Time, t (s)
C2H4
CH4
Run 500.725% POSF10289 (A3) in ArT5 = 1297 K, p5 = 12.72 atm
102
103
104
0.7 0.8 0.9
POSF10289-4%O2-Ar
Igni
tion
Del
ay, s
)
1000K/T
101
102
103
104
0.7 0.8 0.9 1.0
POSF10264-4%O2-Ar
POSF10264, = 1.03±0.09POSF10264, = 0.49±0.01POSF10264, = 2.11±0.06POSF10264, = 1.16±0.19, p5 = 54.4 atm
Igni
tion
Del
ay, s
)
1000K/T
{p5 = 13.0 atm
25
Selected Results – the C1 fuel model
Data: Hanson group Data: Egolfopoulos group
Kinetics for Non - Petroleum Fuel(s) Developed for C1 & C5 (not shown)
102
103
104
0.6 0.7 0.8 0.9 1.0 1.1
Igni
tion
Del
ay, s
)
1000K/T
C1-4%O2-Ar = 1.0, p5 = 15.0 atm
C1-Air = 0.4, p5 = 11.1 atm
C1-Air = 0.9, p5 = 12.1 atm
C1-21%O2-Ar = 1.1, p5 = 15.8 atm
C1-Air = 1.2, p5 = 36.8 atm
C1 (POSF11498) Ignition Delay Time
20
30
40
50
60
70
0.6 0.8 1 1.2 1.4Equivalence Ratio,
T0 = 403 K, p = 1 atm
Lam
inar
Fla
me
Spee
d, S
uo (cm
/s)
C1 (POSF11498) Laminar Flame Speed
26
C1 Ignition Delay Time vs. Cat A Fuels
• Experiments show that C1 ignites slower in the 4%O2-Ar mixture (φ=1), but it has a shorter ignition delay time in the air mixture than A1. Model results support the experimental results.
102
103
104
0.8 0.9 1.0
C1, = 0.9, p5 = 12.1 atm
Cat A, = 1.0, p5 = 11.2 atm
Igni
tion
Dela
y, s
)
1000K/T
Fuel-Air
102
103
104
0.7 0.8 0.9
C1, = 1.0, p5 = 15.0 atmCat A, = 1.0, p5 = 13.7 atm
Igni
tion
Del
ay, s
)
1000K/T
Fuel-4%O2-Ar
FAA CENTER OF EXCELLENCE FOR ALTERNATIVE JET FUELS & ENVIRONMENT
Project manager: Mohan Gupta, FAALead investigator: T. Lieuwen
Co-Investigators: J. Seitzman, W. Sun, D. Blunck, T. Lee, F. DryerResearch Engineers: B. Emerson, D. Noble, D. Wu
Students & Post-Docs: N. Rock, I. Chterev, H. Eck, E. Mayhew, S. Hammack, A. Fillo, B. Sforzo
Advanced CombustionProject 27 a, b, c
Area 3
October 13-15, 2015Seattle, WA
Opinions, findings, conclusions and recommendations expressed in this material are those of the author(s)and do not necessarily reflect the views of ASCENT sponsor organizations.
28
Area 3: Advance Combustion TestsRole Within NJFCP
• Develop and demonstrate advanced combustion screening procedures for alternative jet fuels centered on key figures of merit– Blowoff– Ignition– Turbulent flame speed
• Assess sensitivity of fuel chemistry to turbulence• Input data for some numerical models• Interaction with Area 2 chemistry models• Extra benefits: low fuel consumption, sub-atmospheric conditions
• Elucidate the physics of fuel differences
• Obtain detailed combustor measurements to support modeling efforts (Area 4)
• Refine experimental practices for use in referee combustor (Area 6)
29
Area 3: Accomplishments and Takeaways
• Blowoff– Correlated 1230 blowoff points to physical properties– Measured detailed data for modeling groups– Demonstrated improved 2-camera OH PLIF technique– Demonstrated fuel sensitivities & dependence on injector type
• Ignition– Created large database of ignition probabilities for all fuels– Used Area 2 chemistry to construct ignition model– Demonstrated fuel sensitivities and Developed ignition model
• Turbulent Flame Speed– Measured turbulent flame speeds for different fuels, Reynolds
numbers, and turbulence conditions– Demonstrated fuel sensitivities: Fuel sensitivities change for
different turbulent conditions
30-40
-20
0
20
40
60
80
100
120
140
ΔTad
(F) f
rom
A-2
TBH=375 F
C-1
C-5
A-1C-2
C-4
C-3A-2A-3
-40
-20
0
20
40
60
80
100
120
140
ΔTad
(F) f
rom
A-2
TBH=575 F
Pressure Atomizer
Closed Symbols=Pressure AtomizerOpen Symbols=Airblast Atomizer
Area 3: ResultsTask 1. Blowoff
• Completed over 1230 blowoff measurements!
• Studied blowoffdependence on
– Fuel type– Bulkhead
temperature– Pressure– Injector type
• Demonstrated significant fuel sensitivity
• Example: sample of blowoff shown here
Larger Differences Amongst Cat C Fuels Observed vs. Conventional Fuels
C1 – low cetane, narrow boilingC6 – cycloalkaneC2 – bimodal boilingC3 – high viscosityC4 – low cetane, wide boilingC5 – narrow boiling, full fuel
31
Area 3: ResultsTask 1. Blowoff• Correlated blowoff point to physical
properties• Tried correlating to
– 90% boiling point– 50% boiling point– 10% boiling point– Viscosity– H/C ratio– % parrafins– Many others
• Strong correlation for– 90% boiling point– Viscosity
• Example: correlations to 90% boiling point
-60
-20
20
60
100
140
300 350 400 450 500
ΔTad
(F) f
rom
A-2
90% Boiling Point Temperature (F)
TBH=375 F A-2
A-1
A-3
C-1
C-2
C-3
C-4
C-5
-60
-20
20
60
100
140
300 350 400 450 500
ΔTad
(F) f
rom
A-2
90% Boiling Point Temperature (F)
TBH=575 F A-2
A-1
A-3
C-1
C-2
C-3
C-4
C-5Closed Symbols=Pressure AtomizerOpen Symbols=Airblast Atomizer
Blow-Off (PA) Correlates with Physical Property
C1 – low cetane, narrow boilingC2 – bimodal boilingC3 – high viscosityC4 – low cetane, wide boilingC5 – narrow boiling, full fuel
32
Role:
• Evaluate fuel composition effects on forced ignition under repeatable, engine-relevant, model-able conditions
Results:
• Developed & adapted facility for pre-vaporized fuel injection
• Completed ignition probability screening experiments– demonstrated fuel performance variations
• Characterized ignition kernel growth with schlieren and chemiliuminescence imaging
– Fuel differences observed in flame growth rates
0.2 ms 0.6 ms 1.0 ms 1.4 ms6 m/s
Chemiluminescence
Area 3: Results // Task 2. IgnitionVariation in Novel Fuels > Petroleum Fuel (prevaporized)
Relative Ignition Probability
1.5
C1 – low cetane, narrow boilingC2 – bimodal boilingC3 – high viscosityC4 – low cetane, wide boilingC5 – narrow boiling, full fuel
33
Area 3: ResultsTask 3. Turbulent Flame Speed
Fuel & Air Air
Pilot Fuel & Air
Pilot Fuel & Air
Turbulence Generator
Ball bearings prevent jetting
Able to control• Reynolds number• Preheat temperature• Equivalence ratio• Turbulence intensity
Data Processing• Crop measurements• Assume axisymmetric• 2-D median filter is applied• Able transform performed• Fit to Gaussian distribution to locate flame tip• Flame tip to determine height of cone area
Acone
ICCD Camera• No filter• 230 – 1100 nm• Gate width: 0.07 s• 2 Hz
Test Capabilities and Diagnostics Developed
34
Area 3: ResultsTask 3. Turbulent Flame Speed
Turbulent Consumption Speeds for A2 , C1, and C5, Tpreheat = 390F (470 K)Re=5,000_____________________ _
C5
_
A2_
Re=7,500_______
______________ _Re=10,000_____________________ _
Closed symbols- Low turbulence intensity
(estimated, u’/u = 13%)_
Open symbols - High turbulence intensity
(estimated, u’/u = 19%)_ C5
_A2_
C5
_
Observations:• Turbulent flame speed sensitive to different fuels• Turbulent flame speed increases with Reynolds
number• C5 more sensitive to changes in turbulence
intensity• A2 typically has a greater sensitivity to changes in
equivalence ratio
A2_C5
_
A2_C5
_
A2_C5
_
Results Dependent on Fuel and Flow Conditions
FAA CENTER OF EXCELLENCE FOR ALTERNATIVE JET FUELS & ENVIRONMENT
Project manager: Mohan Gupta, FAALead investigator: Suresh Menon, Georgia Institute of Technology
Co-Investigators: Wenting Sun (Georgia Tech), Tianfeng Lu (U. Connecticut)Post Docs & Students: Dr. R. Ranjan, A. Panchal, Y. Gao, B. Majda, Y. Liu
Combustion Model Development and EvaluationProject 28A
Area 4
October 13-15, 2015Seattle, WA
Opinions, findings, conclusions and recommendations expressed in this material are those of the author(s)and do not necessarily reflect the views of ASCENT sponsor organizations.
36
Area 4: Simulations of Experimental StudiesRole within NJFCP
• Establish simulation strategy using Large-Eddy Simulations (LES) to capture fuel sensitivity in experimental tests– Focus on validation and lean blowout (LBO) for two of the NJFCP
fuels: Cat A2 and Cat C5 in the current year one
• Collaborate with Area 2 to develop efficient reduced reaction kinetics for use in LES– Reduced kinetics for all NJFCP fuels: e.g., Cat A2 and C5 fuels– Network modeling to accelerate computations in LES
• Collaboration with Areas 3, 5 and 6 to do LES of the experimental rig for stable and LBO test conditions– Use Area 5 and OEM scaling data to set spray conditions– Simulate Area 3 rig with the pressure atomized fuel injector– Simulate the Area 6 referee rig with air blast atomizer
37
Area 4: Reduced Kinetics for NJFCP Fuels• Accomplishments
– Reduced models developed for Cat A2 & C5• Based on detailed-lumped models from Area 2• Skeletal: 38 species; Reduced: 29 species
– Extended validation in auto-ignition, perfectly stirred reactors (PSR), flame speed, extinction of premixed & non-premixed counterflow flames
– Analytic Jacobian & non-stiff routines generated for efficient LES– Reduced transport models of 15 groups generated for LES
• Program Takeaways– LBO, ignition and relight requires kinetics need to be accurate and
chemical kinetics are key to show fuel sensitivity, which is the essential element of the current NJFCP program
– Reduced kinetics that can mimic the more detailed reactions can be used with reduced cost in large-scale LES
– Reduction of the cost of kinetics will allow OEM to simulate these problems without sacrificing accuracy
38
Perfectly Stirred Reactor (PSR)Validation at LBO Conditions
• PSR extinction study conducted at LBO conditions (Colket et al., AIAA 2012)• Chemistry of small molecules plays important role at LBO conditions• Reduced models agree well with detailed at LBO
LBO conditions Tin, K P, atm Phi
Case 1 394 2.04 0.457
Case 2 450 2.04 0.436
Case 3 394 3.4 0.456
Case 4 450 3.4 0.434PSR extinction curvesCat A2/air, LBO conditions
Solid lines: lumped-detailedDashed lines: skeletalSymbols: reduced (29 species)
Non-premixed counterflowAuto-ignition
Reduced Model Simulates Detailed Model Well
Detailed Skeletal ReducedCase 2:
Tin = 450 Kp = 2.04 atm = 0.435
Detailed Skeletal ReducedCase 4:
Tin = 450 Kp = 3.4 atm = 0.434
Case 2:Tin = 450 Kp = 2.04 atm = 0.435
Case 4:Tin = 450 Kp = 3.4 atm = 0.434
Residence time, ms
Tem
pera
ture
, K0.1 1 10 100
1000
1200
1400
16001000
1200
1400
1600
Igni
tion
Del
ay, m
s
Max
imum
Tem
pera
ture
, K
1 atm
5 atm
30 atm
Cat A2/Air = 1
0.5 atm
1000/T, K-10.6 0.7 0.8 0.9 1.0
0.5 atm
30 atm
5
1
Cat A2/Air = 1
1
10
100
0.1
0.01
0 6 0 8 1 2 4 6 8 10
10 atm Detailed (112 species) Reduced (29 species)
Cat A2/AirNon-premixed counterflowStream 1: 50% Fuel + 50% N2 (in mole)Stream 2: AirT1 = 300 K, T2 = 300 K
1 atm
1 atm10 atm
0.6 1 2 4 8Reciprocal Strain Rate, S
1800
1200
1400
1600
2000
2200
Stream 1: 50% fuel + 50% N2Stream 2: AirInlet temperatures: 300K
39
Area 4: Network Modeling and Kinetics Acceleration
• Accomplishments– A reactor network model developed to assess kinetic
mechanisms and identify key chemical markers for different fuels– Adaptive kinetic mechanism applied in canonical LES code– Significant speed up of kinetics demonstrated using an on-the-fly
reduction method
• Program Takeaways– Even Reduced kinetics is expensive in LES for design studies– Not all reactions occur everywhere and so optimization should
reduce the overall cost without loss of accuracy in critical areas– Need to develop a general strategy for dynamic adaptation of
kinetics to allow application of the approach to general design– Reduction of the cost of kinetics will allow OEM to use this type of
LES for LBO, ignition and relight for their design without sacrificing accuracy
40
Reactor Network Model
Mixing zone Flame zone
Post-flame zone(only part is shown)
Recirczone
Zero-axial velocity
iso-contour
Flame zone
Air
Air Fuel
LES of spray combustion in LDI
Corresponding reactor network model
Developed based on LES results
~28X faster chemistry~7X faster totally~ time limiting step is NOT chemistry
Reduced mechanism Adaptive mechanism
• Not every species is active everywhere • Calculate only the reactive ones• On-the-fly adaptive kinetics (OAK) mechanism in 3D
turbulent problem• Nearly identical results from reduced mechanism
(38 species from Prof. Lu) and OAK• Significant acceleration (~28X faster for chemistry)• LES of test case is 7X faster without accuracy loss• Time limiting step is no longer chemistry • Application to Area 3 to be demonstrated (Year 2)
Acceleration methods developed for CFD Applications
41
Area 4: LES of Spray Combustion in NJFCP test facilities
• Accomplishments– Area 3 rig simulated using reduced kinetics and spray model
based on available data and scaling laws for Cat A2 fuel– LBO studies underway as well as new Cat C5 fuel studies– Area 6 rig simulations established for Cat A2 fuel
• Program Takeaways– LES captures the unsteady spray-flame-flow interactions– Results match reasonably the available data for stable
combustion in the Area 3 rig with pressure atomizer– Cat C5 fuel stable combustion shows very similar results
consistent with observations – Area 6 simulations for air blast atomizer with full rig effusion
cooling underway for stable and LBO conditions
42
LES of Area 3 Rig using Cat A2/C5 Fuels
• Stable and LBO conditions forCat A2 and C5 being simulated
• Current focus is on comparisonwith data from Area 3 forstable case using assumedspray conditions
• All LES cases employ identicalgrid and models except for thereaction kinetics for A2 and C5
• LBO for Cat A2 and C5 being simulated• Area 6 for stable and unstable cases• Results to be reported later in Oct
Results to be compared to experimental results.
Spray Evolution & Iso-Temperature (1800 K)
OH
FAA CENTER OF EXCELLENCE FOR ALTERNATIVE JET FUELS & ENVIRONMENT
Project manager: Mohan Gupta, FAALead investigator: Matthias Ihme, Stanford University
With: Lucas Esclapez, Peter C. Ma, Hao Wu, Pavan Govindaraju
Combustion and Spray Model Development and Evaluation
Projects 28B & 29BAreas 4b and 5b
October 13-15, 2015Seattle, WA
Opinions, findings, conclusions and recommendations expressed in this material are those of the author(s)and do not necessarily reflect the views of ASCENT sponsor organizations.
44
Takeaways
• Developed and integrated computational submodels into LES for characterization of fuel effects in aviation gas turbines– fuel chemistry, liquid fuel evaporation, droplet distribution, effusive
cooling
• Developed mesh, initial conditions, and boundary conditions to computationally simulate referee-rig experiments within the NJFCP
• Assessed LES-modeling capabilities against experiments for various fuels in Area 5 spray experiments and Area 6 referee rig– Model provides spatio-temporal evolution of the droplet composition– Developed LBO modeling methodology for referee rig
• Models are currently being evaluated against detailed experimental measurements
45
Area 4b: Combustion Modeling Role Within NJFCP
Objectives• Evaluate fuel effects on GT-operability with emphasis on Lean Blowout (LBO)• Develop and integrate combustion models in state-of-art simulation code
Liquid-fuel breakup & spray formation
Multicomponentdroplets evaporation Combustion
Experimentalrepresentation:- Complex geometry- Effusive cooling- Boundary conditions
46
Area 4b: Combustion Modeling Role Within NJFCP
Objectives• Evaluate fuel effects on GT-operability with emphasis on Lean Blowout (LBO)• Develop and integrate combustion models in state-of-art simulation code
Area 4/5: High Fidelity Combustion Simulations to Characterize Jet-Fuel Effects on Combustion Stabilization and Ignition
Area 2: Chemical mechanism for reference fuels
(Wang, Stanford)
Area 5: Validation & development of spray
formation methods (Lucht, Purdue)
Area 6: Experiments of referee combustor
rig(Zabarnick, UDRI)
UTRC: mesh-generation of referee rig; complementary
simulations; submodeldevelopment
47
Area 4b/5b: Simulation of Primary Fuel Breakup
Objectives
• Develop and validate multiphase models for prediction of liquid fuel injection
• Assess and improve low-order correlations models
• Determine droplet distribution as input to combustor simulation
Results
• Droplets size-distribution affects fuel/air mixing, fuel deposition and combustion
• Large liquid/gas interface computations are able to reproduce experimental trends
• Simulations provide representative conditions for integrated simulationExp. from Purdue VOF-simulation
Lb ≈ 10 mm
Qualitative Comparisons are Good
48
Area 4b: Multicomponent droplet evaporationObjective
• Evaporation process critical for ignition, pollutant formation or blowout
• Aviation fuels are mixtures of large number of compounds
Develop physics-based evaporation model that relies on group-contribution representation of realistic aviation fuels
Treatment of Multi-Component Effects Represent Vaporization Rates Better
49
Area 4b: LES simulations of fuel effect on LBO
• Integration of relevant submodels into LES (fuel chemistry, droplet distribution, effusive cooling)– Implementation of the effusive cooling model– More than 58% of the overall mass flow rate through effusion
plates substantial effect on cooling
Withouteffusive cooling
Witheffusive cooling
Simulation of Physics Requires Careful Inclusion on Boundary Conditions
50
Area 4b: LES simulations of fuel effect on LBO
• Integration of relevant submodels into LES (fuel chemistry, droplet distribution, effusive cooling)
• Define LBO protocol to assess fuel effects
= 0.10
= 0.15= 0.2
Simulations predict approach to LBO
FAA CENTER OF EXCELLENCE FOR ALTERNATIVE JET FUELS & ENVIRONMENT
Project Manager: Mohan Gupta, FAA
Lead Investigator: Robert Lucht, Purdue UniversityCo-Investigators: Jay Gore, Sameer Naik, Paul Sojka, Purdue University; Nader Rizk,
ConsultantStudents: Timo Buschhagen, Andrew Bokhart, Rohan Gejji. Robert Zhang
Jet Fuels Atomization Tests and Models Project 29A
Area 5
October 13-15, 2015Seattle, WA
Opinions, findings, conclusions and recommendations expressed in this material are those of the author(s)and do not necessarily reflect the views of ASCENT sponsor organizations.
52
Area 5: Atomization TestsRole Within NJFCP
• Perform detailed diagnostic investigations of spray properties (e.g. fuel droplet size distribution, fuel spray break up length, cone angle) for a selected range of alternative fuels and operating conditions.
• Use advanced diagnostics such as phase Doppler particle anemometry (PDPA). Investigate wide range of operating conditions (e.g., fuel temperature, fuel pressure, swirler pressure drop) using the unique Rules and Tools spray test rig.
• Interact closely with Stanford group (Area 5, Project 29B) that is performing advanced spray modeling, UDRI group (Area 6) that is operating the referee rig, and Georgia Tech group (Area 3) investigating fuel effects in combustion.
53
Area 5: Year 1 Results
• Extensive testing of alternative fuels using PDPA and high-speed video imaging. Measurements performed using referee nozzle, also being used in Area 6 referee test rig. Investigated A-2, C-1, and C-5 fuels over wide range of operating conditions: Fuel temperature: -30⁰ F to 60⁰ F Swirler pressure drop: 2% to 6% Pilot fuel pressure drop: 25 psid to 100 psid PDPA performed at numerous radial locations,
selected axial locations
• Spray quality decreases with decreasing temperature
• Swirler pressure drop has most significant effect on droplet size distribution
54
Area 5: Year 1: Experimental System
Measurement Planes
PDPA Receiver
PDPA Sender
Fuel NozzleApparatus
High Speed Video Camera
55
Area 5: Year 1: Sample ResultsFuel effects observed for C5 fuel.
56
Area 5: Year 1: Sample ResultsRepeatable LBO measurements for various days with unusual C1
FAA CENTER OF EXCELLENCE FOR ALTERNATIVE JET FUELS & ENVIRONMENT
Project manager: Mohan Gupta, FAALead investigator: Joshua Heyne, University of Dayton
Co-Investigator: Tonghun Lee, University of Illinois at Urbana-Champaign Students: Robert Stachler (Dayton),
Anna Oldani (UIUC), and Kyungwook Min (UIUC)
Overall Program Integration and AnalysisProject 34 a, b
Area 7
October 13-15, 2015Seattle, WA
Opinions, findings, conclusions and recommendations expressed in this material are those of the author(s)and do not necessarily reflect the views of ASCENT sponsor organizations.
58
Area 7: Subcommittee on Diagnostics
• Monthly meeting to discuss progress in laser diagnostics and prioritize future goals for Areas 3 & 6.
• Provide platform to reach consensus within the experimental and numerical PIs in the NJFCP. Determination of scope and depth of diagnostics effort.
• Change schedule and diagnostics targets based on needs of the PIs and overall NJFCP direction
• Determine data analysis methodology and format. Integrate opinions from the AFRL research staff
• All minutes from discussions posted on KSN
Tonghun Lee (Illinois), Matthias Ihme (Stanford), Vaidya Sankaran (UTRC), Andrew Caswell (AFRL), Joe Miller (AFRL), Amy Lynch (AFRL), Joshua Heyne (Dayton)
59
Area 7: NJFCP Data Collection
• Determination of data collection protocol for NJFCP (sample, right)
• Acquisition of key data from each area PIs every quarter
• Sort through data for integration into the Alternative Jet Fuel Test Database (Project 33)
• Make data available to NJFCP PIs and wider community
Facilitate Archiving of NJFCP Test Data and Dissemination of Information
sample: data collection protocol