alternative fuel test on pw308 engine - nasa · primary test objective: determine the impact of...

Alternative fuel test on PW308 engine

A Collaborative effortNASA, AFRL, ARI, UTRC and P&W

Anuj Bhargava

AerodyneResearch, Inc.

2

Tests performed to investigate impact of alternative fuels

Primary test objective: determine the impact of synthetic fuel on performance and emissionsSecondary test objective

Impact of heated fuel on emissions and performanceSpatial differences in emissions

Measurements performed on PW 308 engineSampling rake with multiple gas and particle probe used to collect sample at engine exhaust and at 50mMeasurement team

NASA: PM and Gaseous emissionsAFRL: PM and Gaseous emissionsARI: Volatile PM emissionsUTRC: Sampling SystemP&W: Performance and test lead

All test objectives achieved

Successful testing of all three fuels (JP8, Synthetic fuel or Fischer Tropsch (FT) fuel, and Blend)

No significant difference in engine performance or gaseous emissions with the three fuelsPM emission reduction with blend and synthetic fuelHeated fuel had higher NOx emissions

Excellent set of data to validate models

4

Agenda

Tests conductedResults

Engine performance and gaseous emissions with the different fuelsDifference in non-volatile PM emissions Volatile PM emissions

5

Agenda

Tests conductedFuels testedSampling systemTest matrix

ResultsEngine performance and gaseous emissions with the different fuelsDifference in non-volatile PM emissions Volatile PM emissions

6

Test setup

7

Composition of synthetic fuel looks different from JP8

5 10 15 20 25 30Time-->

C11

C12

C10C9

C8 5172 F-T Kerosene

Con

cent

ratio

n

5 10 15 20 25 300

C11C12

C13

C14

C15

C16

C17C18

C10

C9

C8

C7

C19

JP-8

8

The three fuels have similar properties

Property JP-8 Blend SyntheticViscosity (cSt) 1.38 1.14 0.96

Specific Gravity 0.8050 0.7734 0.7377Net Heat of Combustion

(Btu/lb) 18,533 18,735 18,960Hydrogen % mass 13.95 14.78 15.71

Particulate contamination (mg/L) 0.71 0.21 0.11Sulphur content (%) 0.123 0.065 0.003

Aromatic content (%) 19 10.1 0.17

Properties of the blend are within the Aviation fuel SpecsSynthetic fuel doesn’t meet Aviation fuel spec

9

Control

Room

Cooling H2 O

Power

208 Single

Phase, 250 A

PW

308

GN2 - Stand

Air - Stand

Gas lines

Rake

Elevated Test Stand

Aerosol lines

50 m Probe

Communication lines

Valve Boxes

AFRL Trailer

NASA (+ARI) Trailer

Test Setup

10

Particle Probe Plumbing designed to minimize particle losses

AFRL

NASA

50 M Probe Vacuum

Pump

MANIFOLD

1” Conduit, 130’

3/4” PTFE, 20’

MFC Dry N2

0 – 100 LPM

FILTER

3/8” PTFE, 10’

3/8” PVC

DILUTION GAS

Valve Box

ARI

3/4” PTFE, 20’

Considerable effort expensed to insure that the sampling methodology follows the recent advancement made in PM emission measurement

Probes

11

Pretest Line Loss Prediction for 50 m probe

Pretest Line Loss Prediction for 1 m probe to NASA instruments

0.00

0.10

0.20

0.30

0.40

0.50

0.60

0.70

0.80

0.90

1.00

10 100 1000

Particle Diameter (nm)

1M-NASA+AFRL

Pretest Line Loss Prediction for 1 m probe to AFRL instruments

Line loss estimation model predicts line losses within acceptable limits

Acceptable limit

12

Full suite of Aerosol Instruments used to characterize emissions

MAAP

LiCor 6251

3022 CPC

EEPS

Sample Manifold

EC/OC Filter

50 m1 m

AFRL

CPC, SMPS,

TEOM, EC/OC,

Smoke, PAH

P, T, RH

UHSAS SMPS w/3010

Aerodyne AMS

heaterSMPS w/3776

3775 CPC

OVERALL• 6 number densities• 5 size distributions• 4 Mass• 1 Composition

13

1 m Sampling Rake Setup

11

10

9

8

6

5

4

3

2

1

7

Blank

Top

Particle Probe

Particle Probe

Particle Probe

Particle Probe

Particle Probe

Gas Probe

Particle Probe

Gas Probe

Gas Probe

Blank

Engine Center Line

Each probe tip separated by 1.25”

0

2

4

6

8

10

Probe 8

Probe 4

Probe 2

Probe 3

Probe 7

Probe 6

Probe 5

Probe 9

Probe 10

14

heated fuel

heated fuel

PW308 Emissions Test Matrix designed to fulfill all objectives

DAY 1DAY 2DAY 3

Elapsed Time (hh:mm)00:00 01:00 02:00 03:00 04:00 05:00 06:00

Engine Pow

er, % 0

20

40

60

80

100JP-8FT Fuel50/50

Heated fuel

Heated fuel

15

Agenda

Tests performedResults

Engine performance and gaseous emissions with the different fuels

Difference in non-volatile PM emissions Volatile PM emissions

16

Engine and emissions data does not indicate major differences between the three fuels

GASEOUS EMISSIONSAt low power

NOx emissions are within instrument measurement capabilitiesLower CO with FT/blend may be due to higher H/C ratio

At intermediate/high powerVery low CO emissions make ratios irrelevant to evaluate differences between the fuelsNo significant difference in NOx emissions

Negligible UHC at all power conditions for both fuelsSO2 emissions indicate Sulfur content of the blend to be around 50% of JP8 while for 100% FT fuel a value of 0.1% indicates contamination~2% fuel flow benefit with 100% synthetic fuel can be attributed to the higher heat content of synthetic fuel

Negligible differences in gaseous emissions & performance as expected due to similarity in the physical properties of the fuels (like heating value, specific gravity)

Fuel flow NOx COLOW (2200) 0.95

INTERMEDIATE (4500) 0.98HIGH (5750) 1.0

0.97NANA

Thrust (Rotor Speed N1)

RATIO -

Blend/JP8

0.9991.000.995

SO2

0.500.540.54

Fuel flow NOx CO0.90

0.970.98

0.97NANA

RATIO –

FT100%/JP8

0.9850.9820.978

SO2

0.050.10.1

17

Agenda


Engine performance and gaseous emissions with the different fuelsDifference in non-volatile PM emissions

For the different fuels (JP8/Synthetic/blend)

Volatile PM emissions

18

‐34%

‐29%

‐19% ‐18%

‐41%‐43%

‐26%

‐19%

‐3%

‐92%

‐76%

‐50%

‐45%

‐31%

‐91%

‐77%

‐55%

‐45%

‐36%

‐100%

‐80%

‐60%

‐40%

‐20%

0%

Idle 30% 65% 75% 85%

Percen

t Change in PN EI Relative to JP

‐8

Engine Setting

FT blend AFRL

FT blend NASA

FT AFRL

FT NASA

Synthetic fuel reduces Particle Number Emission Index

As expected lower PM emissions with synthetic fuel due to its chemical composition (higher H/C ratio and no aromatics/Sulfur)

19

‐16%

‐29% ‐28%

‐22%

‐28%

‐18%

‐13%

0%

‐80%

‐71%

‐65%

‐57%

‐91%

‐72%

‐46%

‐35%

‐28%

‐100%

‐80%

‐60%

‐40%

‐20%

0%

Idle 30% 65% 75% 85%

Percen

t Change in PM M

ass EI Relative to JP

‐8

Engine Setting

FT blend AFRL

FT blend NASA

FT AFRL

FT NASA

Particle Mass Emission Indexgoes down with synthetic fuel

As expected PM emission reductions not proportional to the fraction of FT fuel Greater discrepancy in the measurements between the different teams

20

As expected greater SN changes with 100% Synthetic fuelthan with the blend

‐18%

‐5%‐7%

‐11%

‐44%

‐39%‐41%

‐44%

‐50%

‐45%

‐40%

‐35%

‐30%

‐25%

‐20%

‐15%

‐10%

‐5%

0%

30% 65% 75% 85%

% Cha

nge Re

lative

to JP

‐8 Smoke Num

ber

Engine Setting

FT Blend

FT

21

0 20 40 60 800.0

0.2

0.4

0.6

0.8

1.0

Nor

mal

ized

Num

ber E

I

Power (%)

FT

50:50 blend

"Cold" TSI3775

FT produces >90% reduction at idle, 40% at high power

Measurement uncertainty is lower than PM EI reduction with synthetic fuel

0 20 40 60 80 1000.0

0.2

0.4

0.6

0.8

1.0

Nor

mal

ized

Mas

s E

IPower (%)

FT

50:50 blend

MAAP

22

Particle size distribution for the different fuels at different powers show similar trends

FT fuel makes lower number of particle emissions as well as particle of smaller size

0E+0

1E+6

2E+6

3E+6

4E+6

5E+6

0 20 40 60 80 100 120 140 160Particle diameter, nm

Num

ber c

ount

JP8 - Idle SYN - Idle

JP8 - 30% SYN - 30%

JP8 - 75% SYN - 75%

23

FT fuels provide larger percentage reductions at lower power conditions

FT emissions are small, more difficult to discriminate from background at

lower thrust conditions

Differences at higher thrust conditions are less

0E+0

1E+6

2E+6

3E+6

4E+6

5E+6

0 50 100 150 200Particle diameter, nm

Num

ber c

ount

JP8Blend100% Synthetic

0E+0

1E+6

2E+6

3E+6

4E+6

5E+6

0 50 100 150 200Particle diameter, nm

Num

ber

coun

t

JP8Blend100% Synthetic

(both figures drawn to same scale)

24

Particle mean diameter decreases with increase in FT fuel content

Particle mean diameter along with particle number count decreases from JP8 to Blend to FT fuel

Idle 30% 65% H 65% 75% 85% H 85%Particle M

ean Diameter (n

m)

Engine Setting

Particle Mean Diameter

JP‐8

Blend

FT

Mass variability for synthetic fuel for the different instruments not well understood

Mass EI estimated or measured with different instrumentsEEPS and SMPS measure size distribution and numbers. Mass estimated using volume of particles and density

MAAP and UHSAS measure mass more directly (based on reflectivityindex)

Estimations based on the assumption that all particles generated by the different fuels have same physical/chemical properties

The differences in the observed mass may be due to Particles generated with the different fuels may have different physical or chemical properties

The different instruments have different cut-off size ranges

26

0 20 40 60 80 100

0.0

0.2

0.4

0.6

0.8

1.0

Nor

mal

ized

"Mas

s" E

I

Power (%)

MAAPUHSAS

EEPS

Reduction in Mass EI much more sensitive and variable for the

different instruments

Reduction in number EI very similar for the different instruments

Mass EI data for 100% synthetic fuel sensitive to instrument selection

0 20 40 60 80 1000.0

0.2

0.4

0.6

0.8

1.0

SMPS

Nor

mal

ized

Num

ber E

I

Power (%)

EEPS

CPC 1

CPC

27

Number EIs variability probably because of variations

in instrument sensitivity

Ratio of “Mass” EIs quite variable at low power,

background particles may be a contributing factor

Blend data shows more variability at low thrust while high thrust data similar to FT fuel

0 20 40 60 800.0

0.2

0.4

0.6

0.8

1.0

Nor

mal

ized

"Mas

s" E

I

Power (%)

SMPS

MAAPUHSAS

EEPS

0 20 40 60 800.0

0.2

0.4

0.6

0.8

1.0

CPC

Nor

mal

ized

Num

ber E

I

Power (%)

EEPSCPC

SMPS

28

Measuring PM emissions at 50m is difficult

Wind causes plume to waver back and forth

across 50 m inlet probe

CO2 @ CN Counts

50 mProbe

29

Comparison between 1m and 50 m data reveals useful information on PM formation

FT fuel produced significantly lower volatile PM emissions at 1m and 50m

0 20 40 60 80 100

0

10

20

30

40

50

Nor

mal

ized

Num

ber E

I

Power (%)

FT50/FT1

Emissions at 50 meters"Cold" TSI3775

J50/FT50

J50/J1

J1/FT1

30

Agenda


Engine performance with the different fuelsDifference in non-volatile PM and gaseous emissions

Between hot and cold fuel


31

Heated fuel (600 F) gives slightly better SFC

0.96

0.965

0.97

0.975

0.98

0.985

0.99

0.995

1

1.005

FNR2 WFR2 TSFC

Performance property

Rat

io h

eate

d/un

heat

ed

JP8-65% JP8-85%Synthetic-65% Synthetic-85%Blend-65% Blend-85%

Heated fuel led to higher combustor inlet temperature and hence improved SFC by 0.5-1%

32

Heated fuel produced more NOx and slightly higher SN

SN differences between the heated fuel and cold fuel was on the order of +10% which is within measurement uncertaintyNOx for heated fuel was ~15% higher for heated fuel

Combustor (nozzles) are not optimized for heated fuelHigher stoichiometric flame temperature with heated fuel

33

Differences between heated and unheated fuels within experimental uncertainty

0.80

0.85

0.90

0.95

1.00

1.05

1.10

1.15

1.20

coldCN hotCN EEPS coldSMPS hotSMPS

Instrument : Number

ratio

: hea

ted/

unhe

ated

JP8 65%JP8 85%Synthetic 65%Synthetic 85%Blend 65%

0.80

0.85

0.90

0.95

1.00

1.05

1.10

1.15

1.20

EEPS-m coldSMPS hotSMPS UHSAS-m MAAP-m SN

Instrument : Mass

ratio

: hea

ted/

unhe

ated

JP8 65%JP8 85%Synthetic 65%Synthetic 85%Blend 65%

PM emission measurements indicate no significant difference between heated fuel and unheated fuel

34

Agenda


Engine performance with the different fuelsDifference in non-volatile PM and gaseous emissions

Spatial variation


35

Spatial variation in the data captures the partial mixing of core and bypass stream

Partially mixed exhaust leads to variation in actual

emissions but EI’s are consistent across the exhaust

0

5000

10000

15000

20000

25000

30000

0 2 4 6 8 10 12

Radius, inch

CO

2, pp

m

0

20

40

60

80

100

120

NO

x, pp

m

CO2NOx

0

1

2

3

4

5

6

7

8

9

10

0 2 4 6 8 10 12

Radius, inch

NO

x EI

36

Just like gaseous emissions, PM number EI indicate little spatial variation

SMPS

0 2 4 6 8 10distance from center

tip 1- 65%tip 2- 65%tip 3 - 65%tip 4 - 65%

CN counter

EEPS

CN counter

0 2 4 6 8 10distance from center

tip 1- 65%tip 2- 65%tip 3 - 65%tip 4 - 65%

EEPS

JP8 SyntheticInstrument-instrument variability much

higher then spatial variation

37

PM emissions with different instruments may not give same answers if instrument size cut-off is different

Differences in total number count between different instruments can be attributed to differences in instruments “cut-off” sizes

0 5 10 15 20

CNCSMPSCNC 1SMPS-1

RUN #TOTA

L PA

RTI

CLE

CO

UN

TS, #

1E+1

1E+3

1E+5

1E+7

1E+9

1 10 100 1000Size, nm

Num

ber o

f par

ticle

s

Two SMPS with different size cut-offs

38

• Negligible Thrust and Fuel Flow impact of FT fuel as compared to JP8

• No significant difference in Gaseous emissions for the different fuels

• PM emissions for the different fuels• Pure FT EI values are an order of magnitude lower in both number & mass at idle

relative to JP8. Differences between the fuels diminish with increasing power.

• Changes in observed Number EIs largely independent of instrument; “mass” EIs sensitive to measurement technique.

• FT particles much smaller, appear to be more dense than JP8

• Emission reductions not proportional to the fraction of FT

• FT suppresses volatile aerosol formation in plume

• At the 50 meter probe, JP8 idle Number EIs 45 times higher that FT; differences decrease with power.

Summary of Non-volatile PM

39

• Spatial variation in emissions indicates unmixedness. Constant Emission EI across the exhaust indicates all emissions vary in a similar fashion

• Useful information to determine mixer efficiency

• Heated fuel had a higher NOx EI than unheated fuel

•

Excellent set of data to model/develop combustion kinetics and impact of fuel composition, thrust (temperature/pressure) on PM emissions

•

Consistent trends with changes in fuels (JP8 vs. blend vs. 100% FT fuel) can be used to resolve the impact of fuel composition on PM emissions

•

Heated unheated fuel data can be used to model impact of temperature on gaseous as well as PM emissions

•

Data collected at different power level can be used to model impact of f/a, temperature and pressure on emissions

Summary of Non-volatile PM (cont.)

40

Agenda


Engine performance with the different fuelsDifference in non-volatile PM and gaseous emissions Volatile PM emissions

41

Exit Plane Organic, Oil & Sulfate EmissionsPM Organic PM Sulfate

EIm

-organic > EIm

-sulfate > EIm

-oil consistent with previous field measurementsVery little volatile PM emission concentration measured at engine exhaust

1.6

1.4

1.2

1.0

0.8

0.6

0.4

0.2

0.0

EIm

-org

anic

(mg

kg-1

)

80604020

Power %

JP-8 100% FT 50% FT

1.6

1.4

1.2

1.0

0.8

0.6

0.4

0.2

0.0

EI m

-SO

4 (m

g kg

-1)

80604020

Power %

JP-8 100% FT 50% FT

0.4

0.3

0.2

0.1

0.0

EIm

-oil

(mg

kg-1

)

80604020

Power %

JP-8 100% FT 50% FT

PM Oil

42

Detailed Spatial Profile of EIM -organic made at engine exhaust

-4

-3

-2

-1

0

1

2V

eriti

cal P

ositi

on (i

nche

s)

86420

Horizontal Position (inches)

JP-8 85% powerFull scale is 0.02 to 2 mg kg-1

Error bars scale with measurement uncertaintyMarker size scales with EIM-organic (ranges from 0.8 ±

0.14 to 1.3 ±

0.4 mg kg-1)

Organic EIs are relatively constant across

the engine exit plane

43

Re-scaled Profile of EIM -sulfate

JP-8 85% powerMarker size scales with EIM

-sulfate (ranges from 0.06 ±

0.01 to 0.40 ±

0.06 mg kg-1)Full scale is 0.02 to 0.2 mg kg-1

Error bars scale with measurement uncertainty

•Plot shows that entrained bypass air is the source of sulfate at 1m

•Signal too weak to generate particle size distribution with AMS

-4

-3

-2

-1

0

1

Ver

itica

l Pos

ition

(inc

hes)

86420


44

Detailed Profile of EIM -oil

JP-8 85% powerFull scale is 0.02 to 2 mg kg-1 Marker size scales with EIM

-oil (ranges from 0.036 ±

0.006 to 1.9 ±

3.3 mg kg-1)Error bars scale with measurement uncertainty

Oil EIs are consistent with an intermittent

source at the periphery

-6

-4

-2

0

2

Ver

itica

l Pos

ition

(inc

hes)

86420


45

Organic PM Emissions at 50mEIm

-

organics ≈

25 ±

10 mg kg-1

• EIorganics

only a small fraction of total PM emissions• Significant amount of lubrication oil (>75%) • EIorganics

independent of thrust condition/fuel

7

6

5

4

3

2

1

Org

anic

PM

(μg/

m3 )

4:17:45 PM3/26/2008

4:18:00 PM 4:18:15 PM 4:18:30 PM

dat

600

550

500

450

400

CO

2 (pp

m)

500x103

400

300

200

100

CPC

(cm

-3)

100% FT, 85% Thrust

Date and Time

Unusual level of lube oil may be due to test engine configuration

46

Organic at 50m is larger than soot

2 3 4 5 6 7 8 9100

2 3 4 5 6 7 8 91000

Dva (nm)

4

3

2

1

0

(dμg

/m3 /lo

gDp)

soot size range

organic PM - plume organic PM - ambient sulfate PM - ambient/plume

100% FT, 85% power

soot coating?

Similar results found at other power conditions and for other fuelsOrganics at 50m have same particle size distribution as periphery of engine exit planeOil is not

coated on the soot –

consistent with emission as liquid droplets

47

EIm

-sulfate (JP8) = 3.5 ±

1 mg kg-1

(1200 ppm sulfur)EIm

-sulfate (50% FT) = 1.5 ±

0.5 mg kg-1

(600 ppm sulfur)EIm

-sulfate (100% FT) < 0.05 mg kg-1

(30 ppm sulfur)

JP-8 – CO2 /number peaks coincide with sulfate peaks

Plumes for JP8 shows contribution of sulfate particles

3.0

2.9

2.8

2.7

2.6

2.5

AM

S S

O4

PM

(μg/

m3 )

4:45 PM3/25/2008

4:46 PM 4:47 PM 4:48 PM

dat

1.0

0.8

0.6

0.4

AM

S O

rgan

ic P

M (μ

g/m

3 )3.0

2.5

2.0

1.5

1.0

0.5

0.0

AM

S O

il PM

(μg/

m3 )

550

500

450

400

350

CO

2 (p

pm)

700x103

600

500

400

300

200

100

CP

C

0.355

0.350

0.345

0.340

0.335

0.330

0.325

NH

4/S

O4

48

PM Plumes for 100% FT fuel shows sulfate concentration less than background

100% FT – CO2 /number peaks coincide with sulfate valleys, i.e. Sulfate concentration in plume is less than ambient sulfate levels

1.90

1.85

1.80

1.75

1.70

1.65

AM

S SO

4 PM

(μg/

m3 )

3:46:00 PM3/26/2008

3:46:30 PM 3:47:00 PM 3:47:30 PM 3:48:00 PM 3:48:30 PM 3:49:00 PM

dat

2.0

1.5

1.0

0.5

AM

S O

rgan

ic P

M (μ

g/m

3 )

7

6

5

4

3

2

1

AM

S O

il PM

(μg/

m3 )

700

600

500

400

CO

2 (p

pm)

600

500

400

300

200

100

x103

49

PW308 Sulfate PM data shows that sulfate PM emissions relate directly to fuel sulfur

PW308 sulfate data follows the trend seen in previous tests

5

4

3

2

1

0

EIm

-sul

fate

(mg

kg-1

)

3.02.52.01.51.00.50.0

Fuel Sulfur Content (g kg-1)

CFM56 Engines N817NA base sulfur (-2C1) N817NA high sulfur (-2C1) N353SW (-3B1) N695SW (-3B1) N14324 (-3B1) N70330 (-3B1) N429WN (-7B22) N435WN (-7B22)

best-fit line R2 = 0.89SO2 = SO3 conversion 0.09 ± 0.1%

Other Engine Types N75853 (RB211-535E4-B) N74856 (RB211-535E4-B) N729FD (PW4158) N616NA (CJ6108A) PW308 AE3007

≥85% thrust

EIm-s

ulfa

te (m

g kg

-1)

Fuel Sulfur Content (g kg-1)

50

Preliminary Data InterpretationVolatile content of PW308 “organic particles” is lubrication oil, partially burned fuel

Lubrication oil emitted as 100-400 nm droplets in the bypass flow – EIm-oil not a strong function of power or fuel

Lubrication oil may be the dominant source of PM especially at low power conditions and especially for synthetic fuels

Partially burned fuel characteristic of hydrocarbon

At engine exit plane, partially burned fuel is coated onto soot

At 50 m, EIm-sulfate depends linearly on fuel sulfur content

At 50 m, plumes generated by FT combustion contain “less”particle sulfate than ambient air

51

WOULD LIKE TO SAY THANK YOU

TO THE WHOLE MEASUREMENT TEAM

alternative fuel test on pw308 engine - nasa · primary test objective: determine the impact of...

Documents