Mohan Gupta, Ph.D. Asst. Chief Scientist

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.


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



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


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



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


0.00

0.01

0.02

0 500 1000 1500

Mol

e Fr

actio

n

Time, t (s)

C2H4

CH4


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


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



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


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



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


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



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


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



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


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



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

Documents

Mohan Gupta, Ph.D. Asst. Chief Scientist