NASA Contractor Repoftq, 32 5 5

Gas Turbine Combustor

C. W. Kauffman

GRANT NSG-3045 NOVEMBER 1980

MSA

I

https://ntrs.nasa.gov/search.jsp?R=19810003595 2018-06-19T16:17:01+00:00Z

TECH LIBRARY KAFB, NM

Il~llllllnlIl~%l~lllIuyI llObL?5b

NASA Contractor Report 3355

Effect of Ambient Conditions on the Emissions From a Gas Turbine Corn bustor

C. W. Kauffman University of Cincitzuati Ciizcinizati, Ohio

Prepared for Lewis Research Center under Grant NSG-3045

National Aeronautics and Space Administration

Scientific and Technical Information Branch

1980

I 111111111 lmlll llIimlmnmnmllL~--. -_----.--- .,.. iiiixn?ri-,-n=, lil I mm-..--

I I

TABLE OF CONTENTS

Page .,

INTRODUCTION. .............................

EXPERIMENTAL EFFORT. ........................

Test Apparatus ................ ; ......... Test Conditions ......................... Experimental Results. .....................

Unburned Hydrocarbon Emissions ............ Carbon Monoxide Emissions ................ Oxides of Nitrogen Emissions. ............. Nitrogen Dioxide Emissions ............... Empirical Emission Correlations. ...........

Discussion .............................

ANALYTICAL EFFORT ..........................

Model .................................

Carbon Monoxide Model. .................. Unburned Hydrocarbon Model ............... Analytical Results .....................

EXPERIMENTAL AND ANALYTICAL COMPARISONS ..........

CONCLUSIONS ..............................

REFERENCES.. .............................

TABLES ..................................

FIGURES .................................

1

4

4 4 5

5 8

10 12 13

14

16

16

1'; 19

26

29

31

33

38

iii

INTRODUCTION

The 1979 Environmental Protection Agency aircraft emissions

standards require that newly manufactured aircraft gas turbine

engines be tested for compliance with emission standards. These

standards are based upon engine testing at standard day condi-

tions. It is extremely difficult to provide the enormous quan-

titiesof air necessary to operate large turbo-fan engines at

standard day conditions, so preliminary emission certification

testing commenced utilizing ambient air. A trend appeared that

the engine to engine variations in emissions were greater than

were thought to be attributable to manufacturing tolerances.

It was postulated that this engine to engine variation was in

part due to the varying atmospheric conditions on the days that

the engines were tested.

The effect of inlet pressure, temperature, and humidity on

the oxides of nitrogen produced by an engine operating at take-

off power was noted quite early, and subsequent correlations

were formulated by Lipfert'. A compilation and evaluation of

these correlations has recently been given by Rubins and 2 Marchionna . For smoke, which is a pollutant of concern at

high thrust settings, a recently reported result3 indicates that

the smoke number variability can be correlated to changes in the

ambient inlet temperature. For a combustor operating at idle

conditions, correlations were developed by Marzeski and

Blazowski' to account for the effects of nonstandard inlet pres-

sure and temperature on gaseous emissions. The effect of ambient

1

temperature and pressure on gaseous emissions over the complete

thrust range for samples of a given production engine have been

correlated by Sarli et al. 5 With the exception of some limited 6 7 engine test data presented by Nelson et al. , Mosier and Roberts ,

and Allen and Slusher 8 , the effect of humidity on idle emissions

has received little attention even though the extreme sensitivity

of CO oxidation to the presence of water vapor is well known.

This investigation was initiated in order to determine, in

a systematic way and under controlled conditions, the effect of

variations in the ambient conditions of pressure, temperature,

and relative humidity upon the emissions of a gas turbine com-

bustor. A single combustor can from a Pratt and Whitney JT8D-17

engine was run at parametric inlet conditions bracketing the

actual engine idle conditions. Nonvitiated inlet air was used,

and fuel was supplied through a flight certified fuel nozzle.

These data were then correlated in order to determine the func-

tional relationships between the emissions and ambient conditions.

These correlations, though useful to account for day to day

variances in emissions, do not give any insight into the mechanism

by which the generation of these pollutants is influenced by

atmospheric conditions. It is for this reason that a mathematical

modelling effort was initiated to explain the observed behavior of

carbon monoxide and unburned hydrocarbons.

The carbon monoxide emissions were modelled using finite

rate chemical kinetics in a plug flow scheme. The combustor was

divided into three zones, primary, secondary, and dilution, and

appropriate values were chosen for the local fuel-air ratio in

2

each zone. The initial species concentrations reflected not

only the local combustor characteristics but also the effect

of the changing inlet conditions.

Hydrocarbon emissions were believed to result from the

failure of large droplets to completely vaporize. 'Vaporiza-

tion calculations were performed for a Rosin-Ranunler fuel drop

distribution as it passed through the three zones of the com-

bustor. The temperatures in each zone were based on the adia-

batic flame temperature for that zone as determined by the

local combustor characteristics and the changing inlet conditions.

3

EXPERIMENTAL EFFORT

Test Apparatus

The experimental program was conducted in a closed duct

combustor test facility described in detail by Fear 9 and located

in the Engine Research Building of the NASA Lewis Research

Center. A single Pratt and Whitney JT8D-17 combustor can, shown

in cross section in Figure 1, was supplied with the appropriate

quantity of Jet A fuel and nonvitiated air. The combustor inlet

conditions of pressure, temperature,and humidity were varied

parametrically around the actual engine idle inlet conditions.

The combustor installation and instrumentation are shown in

Figure 2. The water content of the air furnished to the com-

bustor was controlled by injecting demineralized water into

the hot airstream approximately 5 meters upstream of the com-

bustor. The water was injected through a pressure atomizing

spray nozzle and complete vaporization of the water occurred

before it entered the combustor. The water content of the air

supplied by the preheater was continually monitored using an

EGG optical sensor and it was nominally quite small,with a dew

point of approximately 239 K. The combustor emissions were

measured according to SAE specifications 10 .

Test Conditions

The actual JT8D-17 idle operating conditions are shown in

Table I. Table II lists the parametric inlet conditions investi-

gated in this report. Compressor pressure ratios of 2, 3, 4,

4

and 5 were chosen, with the combustor inlet temperature, T3,

calculated assuming a compressor efficiency of 80%. The mass

flow through the combustor was determined by maintaining a

constant compressor discharge Mach number, M3, or

reference velocity, V3.

a constant

The mass flow into the combustor consists of both air and

water, the combination of which may be considered an oxidizer.

The fuel flow was set to maintain a constant fuel-air ratio

and not a constant fuel-oxidizer ratio. A changing fuel flow

rate was therefore dictated. This resulted in a different

combustor discharge temperature, T4, for each test point. Three

different values of the overall fuel-air ratio were run in order

to investigate the effect of local stoichiometry upon the emis-

sions of the combustor.

Experimental Results

Unburned Hydrocarbon Emissions

The unburned hydrocarbon emission data is shown in Figures

3 to 11. Figures 3 to 7 present data obtained when the combustor

reference velocity was maintained constant at 15.2 meters per

second, and Figures 8 to 11 present data obtained holding com-

bustor inlet Mach number constant at 0.42. Each set of data was

obtained at the four different compressor pressure ratios of

2, 3, 4, and 5. Each figure presents data for one compressor

pressure ratio and contains three sets of data, each corres-

ponding to one of the three overall fuel-air ratios run. Within

each fuel-air ratio grouping the ambient compressor inlet

5

temperature, To, and the relative humidity, RH, were varied.

For each of the three ambient temperatures considered, data

were obtained at three relative humidities with the exception

of the T 0

= 244 K condition where only one value of relative

humidity was run due to extremely small water concentration

necessary to cause saturation.

On all figures the ordinate is the emission index of

unburned hydrocarbons in grams of CH2 per kilogram of fuel

burned, and the abscissa is the combustor discharge tempera-

ture, T4. Early in the program it was thought that the

combustor discharge temperature would be the dominant factor

in determining the emissions.

On all figures a key to the data is presented. The shape

of the symbol indicates the relative humidity, a circle for

zero percent, a square for fifty percent, and a diamond for

one hundred percent. A flag is used to indicate compressor

inlet temperature. No flag indicates 244 K, a vertical flag

289 K, and a horizontal flag indicates 322 K compressor inlet

temperature. The shading of the symbol indicates the overall

fuel-air ratio. Bottom half shading indicates .007, top half

shading indicates . 011, and no shading indicates an overall

fuel-air ratio of .015.

The following trends can beeasily recognized for the

combustor discharge temperature. A higher fuel-air ratio will

result in a higher combustor discharge temperature if all other

conditions are held constant. Increasing the ambient inlet

temperature will result in an increased combustor discharge

6

temperature for a fixed relative humidity and fuel-air ratio.

For a fixed fuel-air ratio and ambient inlet temperature,

increasing the relative humidity decreases the combustor dis-

charge temperature.

The following trends can be recognized for the hydro-

carbon emissions. Curves Al and A7 on Figure 3 show that

an increase in fuel-air ratio results in a decrease in the

hydrocarbon emission index if all other conditions are held

constant. Connection of other points for identical ambient

inlet conditions also gives the same trend. For a fixed fuel-

air ratio, pressure ratio, and zero relative humidity, an

increase in the compressor inlet temperature effects a decrease

in the hydrocarbon emissions as shown by curves B1, B2, and B3

on Figure 3. For a fixed fuel-air ratio, pressure ratio, and

compressor inlet temperature an increase in the relative humid-

ity effects an increase in the hydrocarbon emission index.

This effect is especially noticeable for the 322 K inlet temper-

ature data and is shown by curves C 1, C2, and C3 on Figure 3.

At this condition saturation corresponds to 8.12 mass percent

water vapor. For all conditions fixed except pressure, an

increase in pressure effects a decrease in the hydrocarbon emis-

sion index. In a qualitative fashion one can see that for all

fuel-air ratios the slope of [8(EI-HC)/aTq]RH=0, curves B1, B2,

and B3 of Figure 3,is less negative than the slope of

[a (EI-HC)/BTdRH=l, curves D1, D2 and D3 of Figure 3. For all

curves the slope becomes less negative with increasing fuel-air

ratios. In other words, humidity affects the unburned

hydrocarbon emissions more strongly at lower adiabatic flame

temperatures, and humidity has a stronger influence on emis-

sions than compressor inlet air temperature.

This data clearly shows that the combustor exit tempera-

ture does not uniquely determine the overall emissions as was

first postulated. In order to determine the repeatability of

the data, the experimental conditions run for Figure 3 were

rerun several weeks after the initial data collection. The

results of this run are shown in Figure 4. Although slight

variances in the actual emissions are detected, the qualitative

trends from Figure 3 are clearly reproduced.

Figures 5, 6, and 7 depict the unburned hydrocarbon emis-

sions for compressor pressure ratios of 3, 4, and 5 respectively.

Figures 8, 9, 10, and 11 present the remaining unburned hydro-

carbon emission data taken for a constant combustor inlet Mach

number of 0.42. In all cases data were taken over the same

pressure, inlet temperature, and humidity matrix. The constant

inlet Mach number data displays the same qualitative trends as

the constant reference velocity data.

Carbon Monoxide Emissions

The carbon monoxide emission data is presented in Figures

12 to 20. Figures 12 to 16 present data obtained at a constant

combustor reference velocity of 15.2 meters per second, and Figures

Figures 17 to 21 present data obtained holding combustor inlet

Mach number constant at 0.42.

The test matrix for the carbon monoxide emissions was iden-

ticai with that for the unburned hydrocarbons, as the two were

8

simultaneously measured. On all figures the ordinate is the

emission index of carbon monoxide in grams of carbon monoxide

per kilogram of fuel burned, and the abscissa is the combustor

discharge temperature. A key to the data is presented on all

figures.

The followingtrends can be recognized from the carbon mon-

oxide emission data. Curves El and E7 on Figure 12 show that an

increase in fuel-air ratio results in a decrease in the carbon

monoxide emission index if all other conditions are held constant.

For a fixed fuel-air ratio, Fressure ratio, and zero relative

humidity, an increase in the compressor inlet temperature effects

a decrease in the carbon monoxide emissions as shown by curves

F1, F2, and F3 on Figure 12. F'or a fixed fuel-air ratio, pres-

sure ratio, and compressor inlet temperature an increase in the

relative humidity effects an increase in the carbon monoxide

emission index. This effect is shown by curves Gl, G2, and G3 on

Figure 12. In a qualitative fashion one can see that for all

fuel-air ratios the slopes [>(EI-CO)/aT4)RH=0 , curves F 1' F2' and F 3 on Figure 12, are nearly identical; while the slopes

[a (EI-CwaT41RH=1, curves G 1, G2, and G3, becomes less

negative with increasing fuel-air ratio. At the fuel-air ratio

equal to 0.015 case the slopes are nearly identical. In other

words, changes in relative humidity have a stronger influence

upon the carbon monoxide emissions than do changes in the com-

pressor inlet air temperature.

9

In general the trends shown by the carbon monoxide emissions

are very similar to the trends displayed by the unburned hydro-

carbons. Again one sees that the carbon monoxide emissions are

not uniquely determined by the combustor exit temperature.

In order to determine the repeatability of the data, the

pressure ratio two data was rerun several weeks after the initial

data of Figure 12 was taken. The rerun data is presented in

Figure 13. Although variations in the actual emissions are present,

the qualitative trends are reproduced.

Figures 14, 15, and 16 present the carbon monoxide emissions

for pressure ratios of 3, 4, and 5, respectively. Figures 17,

18, 19 and 20 present the remaining carbon monoxide data taken for

the constant combustor inlet Mach number case. The data was taken

over the same test matrix as for the previous figures presented.

The constant inlet Mach number data display the same qualitative

trends as the constant reference velocity data.

Oxides of Nitrogen Emissions

The oxides of nitrogen data is shown in Figures 21, 22, 23,

and 24 for a constant reference velocity of 15.2 meters per second

and in Figures 25, 26, 27, and 28 for a constant combustor inlet

Mach number of 0.42. The test matrix for the oxides of nitrogen

was identical with that for all other emissions. On all figures

the ordinate is the emission index of total oxides of nitrogen

in grams of nitrogen dioxide per kilogram of fuel burned. The

abscissa is the combustor discharge temperature. A key to the

data is presented on all the figures.

10

For the total oxides of nitrogen, the following trends are

observed. For a given ambient condition the oxides of nitrogen

may either remain constant or increase as shown in Figure 21 by

curves I 1 and I 7' At a given pressure, a fixed fuel-air ratio,

and zero humidity, an increase in the ambient compressor inlet

temperature effects an increase in the emission index, as shown

by curves Jl, J2 and J3 on Figure 21. For a fixed pressure,

fuel-air ratio, and ambient compressor inlet temperature, an

increase in relative humidity causes a decrease in the total

oxides of nitrogen emission index. The effect is quite

noticeable when the amount of water necessary for saturation is

quite large, as shown by curves Kl, K2, and K 3 on Figure 21.

An increase in pressure, with all other parameters constant,

increases the emission index. The slopes [a(EI-NOx)/aT41RH=0,

curves Jl, J2 and J3 on Figure 21, and [a(EI-NOx)/aT41RH=l,

curves L1, L2 and L3 on Figure 21, are seen to be little affected

by overall fuel-air ratio.

Figures 22, 23, and 24 present the remainder of the oxides

of nitrogen emissions for the constant reference velocity case.

Difficulty was experienced with the chemiluminescent instrumenta-

tion during portions of the overall test program. It is for this

reason that incomplete data is presented in Figure 23. Figures

25, 26, 27, and 28 present the oxides of nitrogen emissions for

the constant compressor inlet Mach number case. The constant

inlet Mach number data display the same qualitative trends as the

constant reference velocity data.

11

Nitrogen Dioxide Emissions

The nitrogen dioxide emissions are shown in Figures 29, 30,

31, and 32 for a constant combustor reference velocity of 15.2

meters per second and for compressor pressure ratios of 2, 3, 4

and 5, respectively. Figures 33, 34, 35, and 36 present the

same data for a constant combustor inlet Mach number of 0.42.

The test matrix for-the nitrogen dioxide emissions is identical

with that for all the other emissions. On all figures the ordinate

is the emission index of nitrogen dioxide in grams of nitrogen

dioxide per kilogram of fuel burned. The abscissa is the com-

bustor discharge temperature. A key to the data is presented on

all figures. Figure 31 presents incomplete data due to difficulty

with the chemiluminescent instrumentation.

Most of the nitrogen dioxide emission plots display the same

trends as the oxides of nitrogen emissions. This is shown by the

similarity of the curves on Figure 29 with the curves on Figure 21.

However, for data taken at pressure ratios of 4 and 5, Figures 31,

32, 35 and 36, it is difficult to recognize the functional

dependence of the nitrogen dioxide emissions index on ambient

inlet conditions. An examination of all the nitrogen dioxide

emission data shows that the functional dependence of the nitrogen

dioxide emissions is identical to the total oxides of nitrogen

emissions as long as the combustor discharge temperature is less

than 900 K. Below this temperature the nitrogen dioxide emissions

are a substantial portion of the total oxides of nitrogen. Above

this temperature nitrogen dioxide emissions are no longer a

significant amount of the oxides of nitrogen emissions as the high

temperature causes the nitrogen dioxide to form nitric oxide.

12

Empirical Emissions Correlations

For regulatory purposes the convenient independent variables

in a correlati.on equation are the pressure, temperature, and

humidity at the compressor discharge plane-p3, T3, and HUM,

respectively-and the overall engine fuel-air ratio, FAR. The

experimental data from this study was employed to generate such

an equation for the emission index of each pollutant species.

Because of the similarity between the constant velocity and

constant Mach number data, separate correlations were not developed.

The emission data were fit employing a stepwise multiple linear

regression program to determine the coefficients in an equation of

the following form:

EI = (p3/6.894 x 103) a

exp [b + (FAR/c) + (9T3/5d) + (HUM/e) 1

where the respective dimensions are:

EI(gms/kg), p3(pascals), T3(K), and IlUM(gms I12@/kg air).

The coefficients as determined by the prograr, are given in Table

III for two extremes: all data collected, and various subcases

selected to maximize the correlation. To maximize the correla-

tions for the hydrocarbons and the carbon monoxide, the data

collected at a compressor discharge pressure of two atmospheres

was not included. The correlation consistently under-predicted

the emissions at this pressure ratio due to marginal combustion

in the burner can. For the case of total oxides of nitrogen

only the data for a fuel-air ratio of -015 was included since

the prociuction of oxides of nitrogen is highest under these

conditions.

13

Graphically the agreement between the measured emissions

and the predicted emissions by regression analysis is shown in

Figures 37, 38, and 39 for the selected data case. The rela-

tionship between the emissions and the chosen variables would

appear to be adequately established as the correlation coeffi-

cients, 2 R , clearly show.

Discussion

For a T-56 combustor the effect of ambient conditions on the

gaseous emissions is surprisingly similar to that of the JT8D-17.

The emission data of Marzeski and Blazowski4, shown in Figures 40,

41, and 42, was collected using a T-56 combustor employing two

different fuels, JP-4 and JP-8. Three different geometric modi-

fications were run giving different primary zone fuel-air ratios

of nominal, rich, and lean while maintaining a constant overall

fuel-air ratio. The relative humidity of the inlet air was close

to zero, the simulated compressor pressure ratio was three, and

the compressor discharge Mach number was held constant. The

changing combustor discharge temperature represents changes in

ambient inlet temperature between 244 K and 322 K.

Although the absolute values of the emission indices for the

T-56 vary slightly from the JT8D-17, at identical compressor

pressure ratios and combustor discharge temperatures, the sensi-

tivities dthe emission index to changes in combustor discharge

temperature, i.e. the slopes [a (EI)/aT4]RF,0 o, for hydrocarbons, I . carbon monoxide, and oxides of nitrogen are nearly identical.

Perhaps this is not su.rprising considering that these two com-

bustors are of a similar vintage, loading, air splits, and

14

reference velocity. Such a similarity among various combustors

could ease the regulatory task of developing correction factors

for each combustor for nonstandard inlet conditions.

15

-

ANALYTICAL EFFORT

Model

The experimental results indicate that the hydrocarbon and

carbon monoxide emissions are decreased by increasing fuel-air

ratio, pressure ratio, or ambient compressor inlet temperature;

while they are increased by increasing relative humidity. For

the oxides of nitrogen emissions the situation is just the reverse.

The behavior of the oxides of nitrogen emissions have been modelled

to account for all the observed effects byBlazowski et al. 12

Some of the mechanisms of carbon monoxide production within the

gas turbine combustor have been previously modelled by Morr et al. 13

A less sophisticated model, but including a limited effect of

ambient conditions, has been presented by Sarli5.

In the model considered here and discussed in detail by 14 Subramaniam , it is suggested that the combustor may be treated

as a plug flow reactor in which there is homogeneous reaction

between perfectly mixed fuel and oxidizer under the isothermal

conditions corresponding to the adiabatic flame temperature.

Since the kinetics representing the oxidation of a complex hydro-

carbon fuel, such as Jet A, are only poorly understood, methane

was chosen as the fuel for employment in the chemical kinetics

scheme. In computing the carbon monoxide emissions it must be

emphasized that instantaneous and complete evaporation of the

fuel was assumed.

16

Previous hydrocarbon emission modelling conducted by

Marchionna et al. 15 indicated that much of the unburned hydro-

carbon emissions result from the escape of raw fuel. Therefore

it is necessary to consider the vaporization of fuel droplets

as they pass in a plug flow fashion through the combustor for

the determination of the unburned hydrocarbon emissions.

For both pollutant species the calculation process is

initiated by determining the adiabatic flame temperature within

a particular combustor zone using the NASA CEC-71 computer 16 program . The combustor inlet conditions of temperature,

pressure, and water content were the same as those used during

the collection of the data. The fuel-air ratio within the

combustor was determined by the total fuel flow rate and the air

flow splits within the combustor as shown in Figure 1.

Carbon Monoxide Model

In the kinetic scheme for modelling the carbon monoxide

emissions it was assumed that methane was instantaneously mixed

with air and water vapor in the primary zone to obtain the

desired fuel-air ratio within that zone. The mixture was then

allowed to react for a period of time corresponding to the

appropriate primary zone residence time at a temperature which

corresponded to the adiabatic flame temperature. The primary

zone combustion products were then instantaneously mixed with the

quantity of additional air necessary to simulate entrance into

the secondary combustion zone. The mixture was again allowed to

react at the temperature representing the new adiabatic flame

temperature for a period of time representing the appropriate

17

residence time in this zone. This process was again repeated

in the dilution zone.

The methane-air kinetic scheme employed is that given by

Ay and Sichel 17 and listed in Table IV. The second rate constant

in reaction nine is similar in nature to that developed by

Kollrack", but it is an order of magnitude smaller in order to

prevent excessive oxidation of the carbon monoxide. It may be

worthwhile to note that the species HO2 and NO2 are not included

in the reaction scheme.. Simultaneous solution of the rate equa-

tions for all species was done using the NASA GCKP-72 computer 19 program . The initial species composition utilized in this

program differed for each ambient condition and for each combustor

region.

Unburned Hydrocarbons Model

For modelling the unburned hydrocarbons a distribution of

JP-4 fuel droplets was passed through the respective plug flow

zones of the combustor. The amount of vaporization was determined

by the local adiabatic flame temperature and the local residence

time. Limited atomization data exists for the JT8D-17 fuel nozzle

in the open literature. Therefore a Rosin-Rammler droplet size

distribution function was assumed for the spray. This distribu-

tion gives the weight fraction of particles, R, having a diameter

larger than a given diameter, d, i.e. R = exp (- bag). The value

of the parameter indicating the monodisperse nature of the spray,

q, was assumed to be similar to those determined for airblast

atomizers15. The value of the parameter determining the mean

drop size, b, of the spray was obtained by fitting experimentally

measured emission data at one reference condition. To calculate

18

the amount of fuel vaporized, the drop distribution was divided

into small segments and the diameter squared vaporization law

was applied including corrections for convective enhancement,

i.e.

do2 - d2 = Kt ,

K static = (8/pll) (kg/Cpg) Rn (B + 1) I

and

E=C Pg ItT - Tb)/Ll I

K =K cov static (1 + -276 ReS5 Pr*33) .

The Reynolds number, Re, the Frandtl number, Pr, the gas conduc-

tivity, k g'

and the specific heat, C PFT'

were determined by local

conditions in each zone of the combustor. The droplets were

allowed to remain in each zone for their characteristic residence

time. All droplets were assumed to have an initial velocity of

60 m/s and the drag coefficient was parameterized by:

cD = 28/Re' .48 . 85 +

Analytical Results

Because of limited computer access time, calculations were

performed at a compressor pressure ratio of four only.

Values of the adiabatic flame temperature reflecting the

effects of different ambient conditions are shown in Figure 43.

The effect of humidity is to reduce the flame temperature.

Lines of constant fuel-air ratio may be constructed on this

figure using the data in Figure 44. For a given inlet temperature

and relative humidity a fuel-air ratio is chosen and the value

of the equivalence ratio is determined. This process may then

be repeated for different inlet conditions but the same

19

fuel-air ratio. The results presented in Figure 45 relate

the fuel-oxidizer and equivalence ratios.

It has not been clearly established where the cooling air

from a given louver mixes with the main flow within the combustor.

It may mix immediately within the zone in which it is introduced

or it may not mix until subsequent zones. Hence for a given fuel

flow a maximum and a minimum possible fuel-air ratio can be

computed for each of the three combustor zones depending upon

where the film cooling air is assumed to mix. For an overall

fuel-air ratio of . 011 the zone fuel-air ratios are given in

Table V along with an upper and lower temperature and the resi-

dence time corresponding to these local conditions. For the

primary zone a 2 ms residence time is typical while for the

secondary and dilution zones a 4 ms and a 3 ms residence time,

respectively, are more representative.

In the primary zone the carbon monoxide concentration is

very near the equilibrium value as the computational results show

in Figure 46. Using the plug flow reactor approach the effect

of changing ambient conditions on the amount of carbon monoxide

at the end of the primary zone is shown in Figure 47. Here the

correction factor CFCO is defined as the mole fraction of carbonmon-

oxide at standard ambient conditions (To = 289 K, KH = 0%)

divided by the mole fraction of carbon monoxide at nonstandard

ambient conditions. Three different primary zone fuel-air ratios

are considered, but the effects of ambient temperature and

humidity changes are similar for each.

20

An increase intheambient temperature causesan increase in

the carbonmonoxide mole fraction andanincrease in the ambient

humidity causes a decrease inthecarbon monoxide mole fraction.

These effects areprecisely opposite to that observed in the ex-

perimental measurements butarein agreementwiththe burner mea-

surements of Muller-Dethlefs and Schlader 20 . Theseresults are

simply explained by consideringtheeffect of flame temperature

on dissociation ofC02. As the flame temperature increases dueto

increasing temperature or decreasing humidity, the formation of

CO and 0 is favored. Miles also findsthesame inverse ambient

effects whenthe primary zone is treatedasa perfectly stirred

reactor employing aglobalhydrocarbonkineticscheme. (Private

communication from G. A. Miles, Detroit Diesel Allison Div.,

General Motors Corp., Indianapolis, Indiana in August 1976.)

In view of these results, it appears that the kinetics in

the secondary and dilution zones are the ones primarily respon-

sible for the overall effect of the changing ambient inlet con-

ditions. Some results of modelling the complete combustor are

shown in Figure 48. Here the carbon monoxide mole fraction is

plotted against the combustor exit plane adiabatic flame temper-

ature. Changing parameters are indicated in the identical fashion

as done previously with the experimental data-symbol shape

indicates ambient humidity, flag position indicates ambient

temperature, and symbol shading indicates fuel-air ratio. The

fuel-air ratios here though represent those in each zone of the

combustor. In this particular case the fuel-air ratio of the

secondary zone is varied within reasonable limits because of the

cooling air ambiguity. The primary zone fuel-air ratio is taken

21

as .07 and the exit plane fuel-air ratio is .015. The residence

time in each zone is 5 millseconds. The secondary zone with the

lowest fuel-air ratio (0.025) produced more carbon monoxide than a

secondary zone with a higher fuel-air ratio for all other param-

eters constant. This is not surprising for the colder secondary

zone retards carbon monoxide oxidation. Examining the results

for any one of the secondary zone fuel-air ratios, the effect of

changing ambient conditions on the carbon monoxide is evident.

For zero ambient humidity an increase in the ambient temperature

decreases the emissions, curves A, while for a given ambient

temperature an increase in the ambient humidity increases the

emissions, curves B on Figure 48. The slopes, 0 (co) m4i RH=O o . and [a(CO)/aT41EB,1 o, are seen to depend upon the fuel-air ratio . of the secondary zone.

Within each combustor zone for each operating condition the

local fuel-air ratio is determined by the airflow splits and the

overall fuel and air flows. As the exit plane fuel-air ratio is

increased, the fuel-air ratio within each combustor zone will

also increase. Figures 49, 50, and 51 show the calculated carbon

monoxide emissions at an increased primary fuel-air ratio for

the JT8D-17. The individual secondary zone data pairs with

individual dilution zone data. The values of the fuel-air ratios

used are representative of the ranges which may occur within the

JT8D-17. As expected each figure shows that a higher exit fuel-

air ratio produces less carbon monoxide. Also, within each

dilution sequence the effect due to changing ambient conditions

is discernible. Taking as a reference point the carbon monoxide

emissions at T 0

= 322 K and 0% relative humidity, for all cases

22

considered, an increase in the humidity or a decrease in the

temperature causes an increase in the carbon monoxide emissions.

The slopes [a(CO)/aT41RH=0.0 and [a(CO)/aT41RH=1.0 depend upon

the particular primary zone and the dilution zone fuel-air ratios.

In all cases the calculated emissions are more sensitive to changes

in temperature than to changes in humidity.

For lean idle operating conditions corresponding to a com-

bustor exit fuel-air ratio of 0.007 a slight modification of the

modelling scheme was necessary. Consideration of only three

combustor zones produced very little carbon monoxide at the com-

bustor exit plane. An analysis of the problem indicated that

too small a quantity of carbon monoxide was being produced in

the homogeneous lean primary zone for fuel-air ratios varying

between 0.30 and 0.45. In a lean actual combustion with a fuel

spray I combustion will occur at approximately stoichiometric in

the droplet diffusion flame of the primary zone. These products

of combustion will then be further diluted by the excess air

present in the primary zone. This has the effect of "freezing"

the high carbon monoxide levels which occur during droplet

combustion.

ExperimentalevidencegivenbySullivanindicatesthatthis

type ofprimary zonequenchingdoesoccur. (Private communication

fromD. Sullivan, GeneralElectric Co., Schenectady, NewYork in

March1979.) He foundthatunderlean combustion conditions the

exactplacementofairholesinthe secondary zone hadverylittle

effectonthe carbonmonoxide emission levels, suggestingthatthese

levels arequenchedbysurroundingprimaryairitselfandthatany

additionalairwouldmakeverylittledifference inloweringthemass

23

fractions of carbon monoxide dueto the additionalamountof air

introduced. Asimilar approachwasemployed inthecurrent homo-

geneous combustion model. The methane and the oxidizer were

allowed to react stoichiometrically for a short period of time

(0.5 ITS). This time was chosen to produce large amounts of

carbon monoxide. Then an initial dilution was allowed to occur

within the primary zone to some lower equivalence ratio where

reactions were allowed to continue for the usual 5 ms. These

products were then exhausted into the usual secondary and dilu-

tion zones. The results of such a calculation are shown in

F'igure 52 for subsequent primary zone fuel-air ratios of .06 and

. 03. Two different dilution schemes are then employed in arriv-

ing at the exit conditions. In this fashion adequate amounts of

carbon monoxide may be produced, and the usual ambient effects

may be recognized.

Ey the selection of typical primary zone and dilution zone

fuel-air ratios, it is possible to simulate the JT8D-17 emissions

at different overall fuel-air ratios, including the effect of

differing ambient conditions. Results are presented in Figure 53

where the effects of different ambient conditions are shown on

earlier dilution sequence presented.

For the hydrocarbon emissions, calculations were performed

at one overall fuel-air ratio of 0.011 using the conditions of

Table V. The effect of three different ambient inlet conditions

on the evaporated mass fraction of the fuel spray at the combustor

exit plane is presented in Figure 54 for drops of different

24

diameters. Because of the effect on adiabatic flame temperature,

cold, wet ambient air is effective in suppressing vaporization.

Through a combination of these results and the previously dis-

cussed Rosin-Rammler droplet distribution function, the total

quantity of unburned hydrocarbon emissions may be calculated.

The results are presented in Table VI. For both static and con-

vectively enhanced vaporization, emissions are increased with

respect to the reference level, 322 K and 0% RR, through either

cooling of the ambient air or an increase in humidity. The

emissions are more sensitive to wide humidity variations than to

wide temperature changes.

25

EXPERIMENTAL AND ANALYTICAL COMPARISONS

Both experimental and analytical results show that for zero

ambient relative humidity an increasing ambient inlet temperature

decreases both the hydrocarbon and the carbon monoxide emissions.

Also an increasing ambient relative humidity increases hydrocarbon

and carbon monoxide emissions for a given ambient inlet tempera-

ture.

In order for increasing relative humidity or for decreasing

ambient temperature to increase carbon monoxide production a

modified CO-OH rate constant had to be used in the analytical

model. It was also found that the analytical carbon monoxide

emissions were extremely sensitive to the fuel-air ratio within

each combustor zone. At this time only reasonable bounds may

be placed on the local fuel-air ratio within each zone. As

already indicated by Morr et al. 13 , a Gaussian distribution should

be considered for the local residence times as well as for the

local fuel-air ratios. This was not done in this study.

A direct comparison between experimental and analytical

results is given in Figure 55. The analytical values show too

large an overall decrease in carbon monoxide emissions with

increased overall fuel-air ratio, and too small a sensitivity to

changing humidity and ambient inlet temperature. Additional com-

parisons are given in Figures 56, 57, and 58 where the carbon

monoxide emissions at standard conditions are divided by those

at nonstandard conditions and plotted as a function of ambient

temperature. Relative humidity is also included as a parameter.

The magnitude of the emission variation is reasonable. However,

26

the kinetic calculations are again unable to predict a suffici-

ently large increase in the carbon monoxide emissions with

increasing humidity. Although no calculations were done for the

T-56 combustor, the experimental data is shown in Figure 59 in

the same format as the previous three figures for the JT8D-17.

For no relative humidity the JT8D-17 shows a greater sensitivity

to temperature changes than the T-56.

For the hydrocarbon emissions, as given in Table VI, the

predicted effect of changing ambient conditions is much less than

those actually observed. However, this disagreement is believed

to be due in part to the changing character of the fuel spray as

ambient inlet conditions vary. The data were obtained at a con-

stant fuel-air ratio. Therefore as the water replaced the air

with increasing humidity, it was necessary to decrease the fuel

flow,which caused poorer fuel atomization. At test conditions

for a pressure ratio of two, the entire combustor fuel flow was

supplied by the primary portion of the duplex fuel nozzle. The

Sauter mean diameter, SMD, of the spray is directly proportional

to the fuel mass flow and inversely proportional to the nozzle

pressure drop. It may be calculated by using a proprietary

correlation. This correlation may also be used when there is

fuel flow through the secondary nozzle as did occur for pressure

ratios greater than two. The results however are less certain

for this case. Calculated Sauter mean diameters are superimposed

upon the emission data in Figures 60-63. Large variations do

occur in its value, and for the case of the primary nozzle only

fueled, the highest values of hydrocarbon emissions do

27

correspond to the largest values of the Sauter mean diameter.

The hydrocarbon calculations presented in Table VI,were done

using a constant value for the Sauter mean diameter.

Although no explicit analytical predictions were made for

the oxides of nitrogen an attempt was made to further collapse

the experimental data through consideration of the combustion

efficiency. On Figures 64-67 combustion efficiency is indicated

for each oxides of nitrogen emission data point. It can be seen

that for similar efficiency values there are still large differ-

ences in the oxides of nitrogen emission index.

28

CONCLUSIONS

Changing engine inlet conditions as produced by variations

in the atmospheric temperature and relative humidity causes the

amount of pollutants emitted by a ,gas turbine to vary. Increas-

ing ambient temperature and decreasing relative humidity cause a

decrease in the hydrocarbon and carbon monoxide emissions and

an increase in the oxides of nitrogen emissions. Conversely,

decreasing ambient temperature and increasing relative humidity

cause an increase in the hydrocarbon and carbon monoxide emis-

sions and a decrease in the oxides of nitrogen emissions.

Increasing the engine pressure ratio or the engine fuel-air ratio

decreases the hydrocarbon and carbon monoxide emissions and

increases the oxides of nitrogen emissions and conversely.

For the JT8D-17 combustor these effects were experimentally

quantized. It was found that the variations in the different

emission indices could be correlated by an equation of the form

EI = (p /6 894 x 103)a 3 - ew [b + (FAR/C) + (9T3/5d)

+ (HUM/e)1 ,

using appropriate constants for each pollutant species.

Analytically, it has been possible to determine the effect

of ambient conditions on gas turbine engine emissions. The

combustor is modelled as a plug flow reactor using an appropriate

kinetic scheme along with the best available information concern-

ing local values of the fuel-air ratio and the fuel spray size.

Through the use of a smaller than generally accepted rate constant

for the CO-OK reaction the magnitude and trend caused by changes

29

in ambient conditions of the carbon monoxide emission index were

successfully predicted. This calculated value was found though

to be quite sensitive to the values chosen for the fuel-air ratio

within the different combustor zones. The magnitude and trend due

to changes in ambient conditions of the hydrocarbon emission index

did not reproduce experimental results. This is due to changes in

the fuel spray drop size distribution which occur in conjunction

with changing ambient conditions and which is not included in

the model.

30

REFERENCES

1. Lipfert, F.W., "Correlation of Gas Turbine Emissions Data," ASME Gas Turbine and Fluids Engineering Con- ference, San Francisco, CA, Mar. 1972.

2. Rubins, P.M. and Marchionna, N.R., "Evaluation of NOx Prediction Correlation Equations for Small Gas Turbines," AIAA/SAE 12th Propulsion Conference, Palo Alto, CA, July 1976.

3. Dieck, R.H., Western, E.E., and Reinhardt, A.H., Jr., "Aircraft Gas Turbine Smoke Measurement Uncertainty Using the SAE IEPA Method," J. of Aircraft, 15, 276, 1978.

4. Marzeski, J.M. and Blazowski, W.S., "Ambient Tempera- ture and Pressure Corrections for Aircraft Gas Turbine Idle Pollutant Emissions," ASME Gas Turbine and Fluids Engineering Conference, New Orleans, LA, Mar. 1976.

5. Sarli, V.U., Eiler, D.C., and Marshall,R.I., "Effects of Operating Variables on Gaseous Emissions," APCA Conference on Air Pollution Measurement Accuracy, New Orleans, LA, Oct. 1975.

6. Nelson, A.W., Davis, J.C., and Medlin, C-H., "Progress in Techniques for Measurement of Gas Turbine Engine Exhaust Emissions," AIAA/SAE 8th Propulsion Conference, New Orleans, LA, Nov. 1972.

7. Mosier, S.A. and Roberts, R., "Low Power Turbopropulsion Combustor Exhaust Emissions," Technical Report AFAPL-TR- 73/36, Vol. 3, July 1973.

8. Allen, L. and Slusher, G.R., "Ambient Temperature and Humidity Correction Factors for Exhaust Emissions from Two Classes of Aircraft Turbine Engines," FAA-RD-76-149, Oct. 1976.

9. Fear, J.S., "Performance of a Small Annular Turbojet Combustor Designed for Low Cost," NASA TM X-2476, 1972.

10. Anonymous, "Procedure for the Continuous Sampling and Measurements of Gaseous Emissions from Aircraft Gas Turbine Engines," SAE ARP 1256, 1971.

11. Roberts, R., Fiorentino, A.J., and Greene, W., "Pollution Technology Program Can-Annular Combustor Engines Final Report," NASA CR-135027, 1976.

12. Blazowski, W.S., Walsh, D-E., and Mach, K.D., "Operating and Ambient Condition Influences on Aircraft Gas Turbine NO, Emissions," J. of Aircraft, 12, 2, 1975. -

31

13.

14.

15.

16.

17.

18.

19.

20.

Morr, A.R., Heywood, J.G., and Fitch', A.H., "Measurements and Predictions of Carbon Monoxide Emissions from an Industrial Gas Turbine," Combustion Science and Technology, 11, 97, 1975.

Subramaniam, A.K., "The Effect of Ambient Conditions on Carbon Monoxide Emissions from an Idling Gas Turbine Combustor," Master of Science Thesis, University of Cincinnati, July 1977.

Marchionna, N., Watkins, S., and Opdyke, Jr., "Turbine Fuel Tolerance Study," Avco Lycoming TR 12090, Oct. 1975.

Gordon, S. and McBride, B.J., llcomputer Program for Calcu- lation of Complex Chemical Equilibrium Compositions, Rocket Performance, Incident and Reflected Shocks and Chapman-Jouguet Detonations," NASA SP-273, 1971.

AY, J.H. and Sichel, M., "Theoretical Analysis of NO Formation Near the Primary Reaction Zone in Methane Combustion," Combustion and Flame, 26, 1, 1976. -

Kollrack, R., "Model Calculations of the Combustion Product Distributions in the Primary Zone of a Gas Turbine Combustor," ASME Winter Annual Meeting, New York, 1976.

Bittker, D.A. and Scullin, V.D., "General Chemical Kinetics Computer Program for Static and Flow Reactions, with Application to Combustion and Shock Tube Kinetics," NASA TN D-6586, 1972.

Muller-Dethlefs, K. and Schlader, A.F., "The Effect of Steam on Flame Temperature, Burning Velocity, and Carbon Forma- tion in Hydrocarbon Flames," Combustion and Flame, 27, 205, 1976.

32

TABLE I

Idle JT8D-17 Combustor Conditions

Nominal Operation

Total Inlet Pressure - 2.47 atm Total Inlet Temperature - 393 K Air Flow - 1.37 kg/set Fuel Flow - 0.0161 kg/set Fuel/Air Ratio - 0.0117

TABLE II Test Operation

Compressor Efficiency - 0.8 Compressor Pressure Ratio - 2, 3, 4, 5 Compressor Inlet Pressure - 1 atm Compressor Inlet Temperature - 244, 289, 322 K Compressor Inlet Relative Humidity - 0, 50, 100% Fuel/Air Ratio - 0.007, 0.011, 0.015 Constant Compressor Discharge Mach Number

M3 = 0.42, or Constant Reference Velocity, V3 = 15.2 m/set

33

TABLE III

Coefficients of Regression Analysis

Emission Index HC co NOx Coeff.

All Data i -1.2833 15.806 -0.9468 11.552 -2.916 0.2547

: -128.93 -0.00400 -243.48 -0.00981 324.67 0.02074

e 43.30 76.39 -59.88 Multiple Cor- .934 . 929 . 824 relation Coef. Squared

Selected Data ;: -1.9130 20.135 -1.1214 13.411 -2.090 0.2552

: -107.35 -0.00341 -196.11 -0.00673 334.44 0.0

e 34.61 77.04 -54.31 Multiple Cor- .953 . 940 . 955 relation Coef. Squared

34

1 M + CH4 = CH3 + H 2 CH4 + OH = CH3 + H20 3 CH4 -I- 0 = CH3 + OH 4 CH4 + H = CH3 + H2 5 CH3 + 02 = HCO + H20 6 CH3 + 0 = HCO + H2 7 HCO + OH = co + H20 8 M + HCO = CO + H 9 CO + OH = CO2 + H

10 H + 02 =o + OH 11 0 + H2 =OH +H 12 0 + H20 = OH + OH 13 H + H20 = H2 + OH 14 H + OH = H20 + M 15 H +H = H2 + M 16 0 +o =02 +M 17 0 +H =OH +M 18 N + 02 =NO +0 19 0 + N2 =NO +N 20 OH +N =NO +H

TABLE IV

Kinetic Scheme of Methane/Air Combustion

and Forward Rate Constants

kc =AT"e -AE/RT (cm3/mole/sec) L

Reaction A

0.20E18 0.28E14 0.20E14 0.69E14 0.20E11 0.10E15 0.10E15 0.20E13 0.56E12 0.85E-14 0.22E15 0.17E14 0.58314 0.84E14 0.40E20 0.15E19 0.40E18 0.53E16 0.64ElO 0.14E15 0.40E14

a

0.0

0.0 0.0 0.0 0.0

0.0

0.0

0.5 0.0

7.0 0.0

0.0 0.0

0.0 -1.0 -1.0 -1.0

0.0 1.0 0.0 0.0

AE

88332.5 4962.5 9210.4

11810.8 0.0 0.0 0.0

28584.0 600.0

13895.0 16554.9

9428.8 18004.0 20048.5

0.0 0.0 0.0

5518.3 6232.9

75231.5 0.0

35

F/A

T 2370 T 1.77

T T

T F

T 2380 T 1.90

T 2430 T 1.97

T 2330 T 2.02

T 2330 1870 1520 1130 920 875 T 2.07 1.74 4.79 3.83 2.69 2.54

TABLE V

Typical Local Parameters, PR=4.0, F/A=.011

Primary Secondary Dilution

High Low High LOW High

. 0709 .0484 .0337 .0193 .0119

2400 1.90

2390 1.90

T,=244K 0% RH 1960 1580 1100 1.51 4.20 3.50 T,=289K 0% RH 2015 1635 1195 1.59 4.40 3.50

50% RH 2000 1605 1175 1.60 4.47 3.54

100% RH 1980 1580 1150 1.61 4.52 3.61 T,=322K 0% RH 2050 1680 1235 1.65 4.52 3~58

50% RH 1960 1595 1180 1.69 4.67 3.66

100% RH

850 2.59

. 0109

820 K 2.48 MS

930 905 2.58 2.44

915 900 2.62 2.45

895 895 2.67 2.45

980 950 2.54 2.45

940 925 2.64 2.46

Low

TABLE VI

HYDROCARBON EMISSIONS

Ambient Correlations Hydrocarbon Emission Index Experi-

Static Convective mental

322 K, 0% RH 1.5 1.5 1.5 244 K, 0% RH 3.19 2.80 14.4 322 K, 100% RH 4.35 3.74 17.6

36

-

FUEL INJECTOR AND PRIMARY SWPRLER EQUIVALENT METERING AREA 7.61%

Equivalent Metering Area

Louver Cooling Air Combustion Air

Panel

1 1.53 2 7.93 2 5.62 3 1.92 3 7.56 5 8.00 4 5.69 8 15.85 5 4.24 9 18.09 6 3.41 7 3.42 8 3.43 9 2.78

10 1.81

% Panel %

Figure 1. JT8D-17 Combustor

37

-

A STATIC PRESSURE o TOTAL TEMPERATURE a GAS SAMPLE PROBE

SECTlON A-A SECTION B-B SECTION C-C

COMBUSTOR INLET THERMOCOUPLE LOCATION GAS SAMPLE PROBE LOCATION

ilNSULAJION-WRAPPEB LINER

Figure 2. Combustor Test Section and Instrumentation Sections

38

0 0.0 RH m NO FLAG TO=244 q 0.5 RH. VT. FLflG TO=289 0 1 .O RH. HOR.FLAG TO=322

;;tlTOH &I;;~; - F/R=.007 - F/R=.011

OPEN - F/A= .015

PR= 2 V3 =CONST.

I O45 -00

I 55 I -00 I 65 000 75.00--- I

m10' 85.00

1 95 .oo

T4 - K

Figure 3. Hydrocarbon Emission Index for a Constant Reference Velocity at a Pressure Ratio of Two

39

0 0.0 RH. NO FLAG TO=244 III 0.5 RH. VT. FLAG TO=289 0 1 .O RH. HOR.FLFIG TO=322

I I 1 I I .oo 55 .oo 65 .OO 75 moo 85 .OO 95.00 14 - K mlOX

Figure 4. Hydrocarbon Emission Index Rerun for a Constant Reference Velocity at a Pressure Ratio of Two

40

0 O’.O RH. NO FLAG TO=244 0‘0.5 RH. VT. FLAG TO=289 0 1 .O RH. HOR.FLAG TO=322

I PR= 3 V3 =CONST l

4!h m- o- 0

e- o- ---- 1 I I

65.00 75 .oo 85 .OO 95.00 105 -00 T4 - K m10'

Figure 5. Hydrocarbon Emission Index for a Constant Reference Velocity at a Pressure Ratio of Three

41

0 0.0 RH. NO FLAG TO=244 tl 0.5 RH. VT. FLAG TO=289 01.0 RH. HOR.FLFIG TO=322

;3;TOM ;;;l-lE; - F/R=.007 - F/A=.011

OPEN - F/F\=eO15

PR= 4 V3 =CONST.

e Q 5 0 o-

I moo 1 65 000 I *& o- 75 .oo 85 -00 I

m10' 95 .oo

1 105 .oo

T4 - K

Figure 6. Hydrocarbon Emission Index for a Constant Reference Velocity at a Pressure Ratio of Four

42

0 0

. W

0 0

d

9

0 0.0 RH. NO FLAG TO=244 00.5 RH. VT. FLAG TO=289 0 1 .O RH. HOR.FLAG TO=322

BOTTOM ;;;iE; - F/A=.007 TOP - F/R=.011 OPEN - F/R= ,015

PR= 5 V3 =CONST.

e a-

%

e-

e ti so0 e- I I I

3.00 70 .oo 80 -00 90 .oo 100 moo lf.O.00 T4 - K x10'

Figure 7. Hydrocatibon Emission Index for a Constant Reference Velocity at a Pressure Ratio of Five

43

z . z!, 4

0 0.0 FiH. NO FLAG TO=244 El 0.5 RH. VT. FLAG TO=289

=: . o- 0 1 .O RH. HOR.FLRG TO=322

I PR= 2 H3 =CONST.

53 e-

. 0 z r!!F

Ei

;45 .oo I I I I I

55.00 65.00 75 .oo 85 -00 95 T4 - K x10'

Figure 8. Hydrocarbon Emission Index for a Constant > Compressor Discharge Mach Number at a

Pressure Ratio of Two

44

=: .

0 -J-- .-a I I

0 0.0 RH, NO FLFlG TO=244 00.5 RH, VT. FLAG TO=289

00 0 1.0 RH. HOR.FLAG TO=322

.

No, d

00 .

EL +

( PR= 3 H3 =CONST. 1

0 0 e- .

0, N

.F- oo-

e-

t2 a?-

45

o-

.oo 65 I .OO -------------- 75.00 85 .OO 9s .oo 205 .~OO T4 - K x10'

J'iqure 9. Hydrocarbon Emission Index for a Constant Compressor Discharge Mach Nutilber at a Pressure Ratio of Three

45

0 0.0 RH. NO FLAG TO=244 El 0.5 RH. VT. FLAG TO=289 0 1 .O RH. HOR.FLAG TO=322

yI;TOH ;;;Fi;E - F/A=.007 - F/A=.011

OPEN - F/A= .015

, PR= 4 H3 =CONST a

0 0 0 0

& & 0 la

o-

0 0 0 0

ci ci e e

0 o- E- E- 0 0

=: =: %

e- c;ss .oo c;ss .oo

I I I I I I I @

65 -00 65 -00 75 -00 75 -00 85 .OO 85 -0 $3

.OO X10' X10'

95 900 95 900 105 -00 105 -00 T4 - K T4 - K

Figure 10. Hydrocarbon Emission Index for a Constant Compressor Discharge Mach Number at a Pressure Ratio of Four

46

I- ~------ ----.---.-.-.-- ----...‘.“--’ ..-- ‘-. ‘.‘. .“’ ”

0 0.0 RH. NO FLAG TO=244 El 0.5 RH. VT. FLAG TO=289 0 1 .O RH. HOR.FLAG TO=322

PR= 5 H3 sCiONSTo

A

O- -e- I I I I

1.00 70 .oo 80.00 90 moo 100 .oo 110 T4 - K m10'

Figure 11. Hydrocarbon Emission Index for a Constant Compressor Discharge Mach Number at a Pressure Ratio of Five

47

0 0.0 RH. NO FLRG TO=244 Cl 0.5 RH. VT. FLAG TO=289 0 1 .O RH. HOR.FLRG TO=322

BOTTOM :;;!-I:; I ;;;=.;y: TOP OPEN - F/R-:015

PR= 2 V3 =CONST.

I I I I I .oo 55 n oo 65 .OO 75 .oo 85 -00 95.00 T4 - K x10'

Figure 12. Carbon Monoxide Emission Index for a Constant Reference Velocity at a Pressure Ratio of Two

48

z . z .

0 0 03 03, d -1 0 0.0 RH, NO FLAG TO=244

00.5 RH. VT. FLAG TO=289

=: 0 1 .O RH. HOR.FLRG TO=322

.

z, 4

PR= 2 V3 =CONST. J

\ \ \

-Y

=: .

O-

*4k. 00 &.oo I I I I

65 .OO 75 .oo 85 .OO 95 T4 - K MO'

5.00 5’5 .oo S’S .OO 7k .oo 8k .OO 9k T4 - K MO'

.

Figure 13. Carbon Monoxide Emission Index Rerun for a Constant Reference Velocity at a Pressure Ratio of Two

49

d 4-l

z .

EL m-4

0 0

.

=3- d

=: .

O- W

0 n-

0 0.0 RH. NO FLAG TO=244 fJ0.5 RH. VT. FLAG TO=289 0 1.0 RH. HORmFLRG TO=322

l33;TOM ;;;;E; r ;;;=.W;

OPEN - F/R::015

PR= 3 V3 XONST.

0 0

‘0 -0

Wd- e UY

0 B-

z .

O- *

e- 4

8 0 cl- 0

.

E o-

0

%5 .oo I I I I I

65.00 75 .oo 85 .OO 95 .oo 105 coo T4 - K do'

Figure 14. Carbon Monoxide Emission Index for a Constant Reference Velocity at a Pressure Ratio of Three

50

8

A

0 0.0 RH. NO FLAG TO=244 q 0.5 RH. VT. FLAG TO=289 0 1.0 RH. HOR.FLAG TO=322

BOTTOM SHADED - F/R=.007 TOP SHADED - F/R=.011 OPEN - F/R=.015

I PR= 4 V3 =CONST. I

e

6 d!P-

0

o-

e-

e- .

I I I I I

5.00 65 -00 75 .oo 85.00 95 -00 105 .oo T4 - K m10’

Figure 15. Carbon Monoxide Emission Index for a Constant Reference Velocity at a Pressure Ratio of Four

51

0 0.0 RH. NO FLAG TO=244 0 0.5 RH. VT. FLAG 0

TO=289 1.0 RH. HOR.FLRG TO=322

BOTTOM SHRDED - F/R=.007 TOP SHADED - F/R=.011 OPEN - F/R=.015

0-I PR= 5 V3 =CONST.

=:

“so.00 I I I --

70 .oo 80 I

.OO 90 -00 m10'

100 .oo 1’10 T4 - K

.co

Figure 16. Carbon Monoxide Emission Index for a Constant Reference Velocity at a Pressure Ratio of Five

52

0 0.0 RH. NO FLRG TO=244 tl 0.5 RH. VT. FLAG TO=289 0 1.0 RH. HORmFLRG TO=322

Id- 8

Q

o-

e- El”

I o- /- ~-

I I I I I 5.00 55 -00 65 .OO 75.00 85 -00 9s.00

T4 - K x10'

Figure 17. Carbon Monoxide Emission Index for a Constant Compressor Discharge Mach Number at a Pressure Ratio of Two

53

8

o- r 0 0.0 RH. NO FLAG TO=244 III 0.5 RH. VT. FLAG TO=289 0 1 .O RH. HOR .FLRG TO=322

BOTTOM SHRDED - F/R=.007 TOP SHRDED - F/R=.011 OPEN - F/R=.015

I PR= 3 M3 =CONST. I

I I I I -a-

“00 65.00 75.00 85.00 95.00 105 -00 T4 - K d0'

Figure 18. Carbon Monoxide Emission Index for a Constant Compressor Discharge Mach Number at a Pressure Ratio of Three

54

0 0.0 RH. NO FLRG TO=244 q 0.5 RH. VT. FLRG TO=289 0 1.0 RH. HOR -FLAG TO=322

+I ;;;TOM ;;;t-K& z ;;;=.;CI;

OPEN - F/R::015

8 I PR= 4 H3 =CONST. I

e-

0 0 0

.

& cc4 R- u

eb 0 0

. P- 4

e-

= .

l-55 .oo I I I I

..-p-

65.00 75 .oo 85.00 95 -00 105. T4 - K do’

00

Figure 19. Carbon Monoxide Emission Index for a Constant Compressor Discharge Mach Number at a Pressure Ratio of Four

55

% 0 0.0 RH. NO FLAG TO=244 q 0.5 RH. .VT. FLAG TO=289

o-‘0 1.0 RH. HORmFLRG TO=322

BOTTOM SHADED - F/R=.007 TOP SHRDED - F/R=.011 OPEN - F/R= -015

I PR= 5 H3 XONST. I

0 0

D 0 o- to- 0 m-l

z z . .

O60 -00 O60 -00 I I

IO0 -00 IO0 -00 I I

70 .oo 70 .oo 80 -00 80 -00 90 -00 90 -00 110.00 110.00 T4 - K T4 - K m1o1 m1o1

Figure 20. Carbon Monoxide Emission Index for a Constant Compressor Discharge Mach Number at a Pressure Ratio of Five

56

z .

01

& x 0 z I

-0

0 0.0 RH. NO FLRG TO=244 tlO.5 RH. VT. FLRG TO=289 0 1 .O RH. HOR.FLAG TOz3.22

;;;TOf’l ;;;tiE~ - F/A=.007 - F/A=.011

OPEN - F/A= .015

PR= 2 V3 =CONST.

J3 ./,

I I .oo 55 l oo 65 -00 -TOT 85 I -00 I

T4 - K m1cl’ 95.00

Figure 21. Oxides of Nitrogen Emission Index for a Constant Reference Velocity at a Pressure Ratio of Two

0 0.0 RH. NO FLAG TO=244 17 0.5 RH. VT. FLRG TO=289 0 1 .O RH. HOR.FLAG TO=322

i33;TOH ;;;;Ei I ;;;==;(I:

OPEN - F/R::015

PR= 3 V3 =CONST l

6

o-

0 cl-

8 8 n- 5+ o- A I I I -1

i .oo 65 -00 75 000 85 -00 X5.00 105 -00 T4 - K d0'

Figure 22. Oxides of Nitrogen Emission Index for a Constant Reference Velocity at a Pressure Ratio of Three

58

-.. . . . .- _.-. -_.... . .-.... -.---.- .__.-.. . . . . -..._ ---.__, _ _ .

0 0.0 RH. NO FLAG TO=244 Cl 0.5 RH. VT. FLAG TO=289 0 1 .O RH. HOR.FLAG TO=322

;i&TOM E;;MWE; I ;;;=SICI:

OPEN - F/A::015

PR= 4 V3 =CONST.

o-

I I I I I i.00 65 -00 75.00 85 .OO 95 .QO 105 .oo

T4 - K d0'

Figure 23. Oxides of Nitrogen Emission Index for a Constant Reference Velocity at a Pressure Ratio of Four

59

0 0.0 RH. NO FLAG TO=244 El 0.5 RH. VT. FLAG TO=289 0 1 .O RH. HORnFLAG TO=322

;ElTOM ;;;liE; - F/A=.007 - F/R=.011

OPEN - F/A=.015

PRr 5 V3 =CONST.

e

8

4

g, 8

O R- I!!!-

o- o-

I I I .oo 70 .oo 80 .OO

I 90.00 100.00 -1110.00

T4 - K m10'

Figure 24. Clxides of Nitrogen Emission Index for a Constant Reference Velocity at a Pressure Ratio of Five

60

.-.---- .- . _--_ ._. . . .._ . . . . . , I II I I

0 0 .O RH. NO FLAG TO=244 q 0.5 RH. VT. FLAG TO=289 0 1 .O RH. HOR .FLRG TO=322

TC$TOM SHROEO - F/fl=.OO7 SHROED - F/A=.011

OPEN - F/R=.015

PR= 2 Pl3 =CONST.

o-

e-

0 b

---. I I ----I------. ‘.OO 55 000 65 .OO 75 .oo 85.00 ---- 95 COD T4 - K X10’

Figure 25. Oxides of Nitrogen Emission Index for a Constant Compressor Discharqe Mach Number at a Pressure Ratio of Two

61

0 0.0 RH, NO FLRG TO=244 00.5 RH. VT. FLAG TO=289 0 1 .O RH. HOR.FLAG TO=322

!3C?;TO” $?f?~~ I ;:‘$=mf

OPEN - F/A::015

PR= 3 II3 =CONST.

b

o-

e+

tl 0

cl-

c- I

o- , 1

75 .oo I

85.00 95.00 105.00 T4 - K X10'

Figure 26. Oxides of Nitrogen Emission Index for a Constant Compressor Discharge Mach Number at a Pressure Ratio of Three

62

0 0.0 RH, NO FLAG TO=244 q 0.5 RH. VT n FLRG TO=289 0 1 .O RH, HOR.FLAG TO=322

BOTTOM SHADED - F/A=.007 TOP SHADED - F/A=.011 OPEN - F/A=.015

PH= 4 H3 =CONST.

P

0 0 8

o- o-

o-

e

0

cl-

-6’s -00 I I

-00 75.00 I I

85 -00 95 .oo 105 .oo T4 - K m10’

Figure 27. Oxides of Nitrogen Emission Index for a Constant Compressor Discharge Mach Number at a Pressure Ratio of Four

63

0 0 l O RH. NO FLFIG TO=244 q 0.5 RH. VT= FLRG TO=289 0 1.0 RH. HOR.FLAG TO=322

BOTTOM ;I-&;;:; - F/A=.007 TOP - F/R=.011 OPEN - F/A=.015

PR= S M3 =CONST.

e-

b

o-

0

II no0 70.00 80 -00 --------I 90.00

T4 - K m10' 100.00 110.00

Figure 28. Oxides of Nitrogen Emission Index for a Constant Compressor Discharge Mach Number at a Pressure Ratio of Five

64

0 0.0 RH, NO FLAG TO=244 00.5 RH. VT. FLAG TO=289 0 1 .O RH. HOR.FLAG TO=322

I I I I .oo 55 -00 65 ..,OO 75.00

I

14 - K ~10’ 85 .OO 95 .oo

Figure. 29. Nitrogen Dioxide Emission Index for a Constant Reference Velocity at a Pressure Ratio of Two

65

0 0.0 RH. NO FLAG TO=244 00.5 RH. VT. FLAG TO=289 0 1 .O RH. HOR.FLAG TO=322

;3;TOH SHFlDED - F/A=.007 SHfXIED - F/A=.011

OPEN - F/FI=.OlS

PR= 3 V3 =CONST.

I B-

e-

b 0

z %s .oo

I 65 -00

1 75 l oo 85 -00 95 .oc lCl5 .OCl

T4 - K x10'

Figure 30. Nitrogen Dioxide Emission Index for a Constant Reference Velocity at a Pressure Ratio of Three

66

0 0.0 RH, NO FLAG TO=244 00.5 RH, VT. FLAG TO=289 0 1 .O RH. HORmFLRG TO=322

o-

I I I I I -00 65 .OO 75 -00 85 -00 95 no0

m10' 105 moo

T4 - K

Figure 31. Nitrogen Dioxide Emission Index for a Constant Reference Velocity at a Pressure Ratio of Four

67

0 0.0 RH. NO FLRG TO=244 Cl 0.5 RH. VT. FLRG TO=289 0 1 .O RH. HOR.FLRG TO=322

;;;TOH ;;;tM& I F;;=Jly:

OPEN - F/R-:015

PR= 5 V3 =CONST a

e c&P St 6

e

0

o- o- o- bo-

I I r 1 moo 70 moo 80 -00

1 90 900

d0' 100 -00 110.00

T4 - K

Figure 32. Nitrogen Dioxide Emission Index for a Constant Reference Velocity of a Pressure Ratio of Five

68

0 0.0 RH. NO FLRG TO=244 00.5 RH, VT. FLAG TO=289 0 1 .O RH, HOR.FLRG TO=322

!33;TOH SHRDED - F/R=.007 SHADED - F/R=.011

OPEN - F/R= .015

PR= 2 H3 =CONST l

6 e

B-

a-

o-

e-

6 0

a-

I - I I I I i .oo 55.00 65.00 75 moo 85.00 95 l oo

T4 - K x10' Figure 33. Nitrogen Dioxide E'mission Index for a

Constant Compressor Discharge Mach Number at a Pressure Ratio of Two

69

0 0.0 RH, NO FLRG TO=244 OO.5 RH. VT. FLAG TO=289 0 1 .O RH. HOR.FLRG-TO=3?2

BOTTOM ;ti;;E; 1 ;;;=.;y; TOP OPEN - F/R::015

PR= 3 H3 =CONST.

8 a-

o-

e-

e

I¶-

e

0

cl-

o-

I I I I 1 -00 65 -00 75 moo 85 -00 95.00

x10' 105.00

T4 - K Figure 34. Nitrogen Dioxide Emission Index for a

Constant Compressor Discharge Mach Number at a Pressure Ratio of Three

70

0 0.0 RH. NO FLAG TO=244 q 0.5 RH, VT. FLRG 0 1.0 RH. HOR.FLRG &I

0~;;;

;;;TOH SHADED - F/R=.007 SHRDED - F/R=.011

OPEN - F/R=.015

PR= 4 H3 =COb$T.

8 4 e

Et

% e-

Q 0

b- 6 b 6%

o-

I I I -------- moo 65 -00 75 l oo 85 .OO 95 .tlo ~05.00

T4 - K x10'

Figure 35. Nitrogen Dioxide Emission Index for a Constant Compressor Discharge Mach Number at a Pressure Ratio of Four

71

0 0 .O RH. NO FLAG TO=244 00.5 RH, VT. FLAG TO=289 0 1 .O RH. HOR.FLAG TO=322

BOTTOM SHRDED - F/R=.007 TOP SHADED - F/R=.011 OPEN - F/R= .015

PR= 5 H3 =CONST.

8

e

6 b

e-

I I -.-.. -- _- I

.oo 70 moo 80 -00 90.00 T4 - K x10"

----zcoo i'm .QO

Figure 36. Nitrogen Dioxide Emission Index for a Constant Compressor Discharge Mach Number at a Pressure Ratio of Five

72

MEASURED HC EMISSION iNDEX

Figure 37. Hydrocarbon Emission Curve Fit

73

MEASURED CO EMISSION INDEX

Figure 38. Carbon Monoxide Emission Curve Fit

74

MEASURED NOx EMISSION INDEX

Figure 39. Oxides of Nitrogen Emission Curve Fit

75

0 JP-4 •I JP-8 0 JP-4 lean A JP-8 lean fL JP-4 rich

“550 650 750 T4-K

Figure 40. Hydrocarbon Emission Index, T-56

76

92

82

72

62

42

32

0 JP-4 0 JP-8 0 JP-4 lean A JP-8 lean b JP-4 rich

i0 650 750 850 950 1050

T4 - Cc Figure 41. Carbon Monoxide Emission Index, T-56

77

4 2 .Y

38 .

33 .

% 28 .

0 3 L 23 h .

18 .

13 .

. 8

0 JP-4 •I JP-8 0 JP-4 lean A JP-8 lean b JP-4 rich

550 650 750 850 950 IO50

T4- K

Figure 42. Oxides of Nitrogen Emission Index, T-56

78

2500

2300

2100

1900

1700

I500

I300

I100

900

t

I-.! 700

.

z 500

5 lu

PR=40

To= 322K

2 2300 -

F

8 2100 -

z rl 1900 -

cu

u -r-l 1700 - -lJ

zi 4 I500 -

2

1300 -

/ RH’IL) I

I 500

700 I- ’ I I I I J .2 .3 4 .5 .6 .7 .0 .9 1.0 I.1 l2 13

Equivalence Ratio, @

Figure 43. Methane Adiabatic Flame Temperature

79

-- I 1, =322K

!?H=O.O M

9 ao- .O 9

.Oi I 1 I mu,nn 1 / 1

/// / -1 n-4

5

/

/ a I t

“n’-“-” H / I 5 .“W 1 RH- 1.0 /

T,, = 24416

I I I I I I I I I I 1 J .2 .3 4 5 .6 7 .6 .9 1.0 1.1 12 I3

4

Figure 44. Methane Fuel-Air Ratio

80

*r

.07 -

D6-

04 -

03-

.02-

.Ol -

.W- s .08 -

.07 -

.06 -

.05 -

.04 -

.03 -

.02 -

.oi -

.oo I I I I I I I I I I I 1

.2 .3 4 .5 .6 .7 .6 .9 1.0 I.1 1.2 1.3 a

Figure 45. Methane Fuel-Oxidizer Ratio

81

.I0

. 09

. 08

. 07

,.06 -

. 05 -

. 04 -

. 03 -

. 0% - I

C

.Ol -

o-

. 06

Equilibrium ----=--- Kinetic, Tp=5ms

I I I 1

.07 .08 .09 .to

Figure 46. Primary Zone Carbon Monoxide Mole Fraction

82

P.R.=4,~p=Sms, CF ~0 = ppm std/ppm non-std

(f/alp =.O 8

a-.-- R H = 0 %

I- (f/alp =.O 7

RH=lOO%

05&- 210 r

1.5 t 8

1.0 - RH=O%

0.5 ’ I I I I I I I 1 .I 240 260 280 300 320

To-K Figure 47. Normalized Primary Zone Carbon Monoxide Emissions

83

24- 0 0% RH; WI flag-To=244 K (f/a)p=o7 o 50%RH; vt. f lag-To=278 K (f/a)s= 035 btm.shade

22 - 0 IOO”/~RH; hor. f lag-To=322K .030rt. shade 0275 top shade

20- 025 Ift. shade (f/a)d = 915

Q 18-

Q 2 ‘6-

4-

2- A

01 I I I I I I I I I 600 800 IO00 1206 1400

T4 - K Figure 48. Carbon Monoxide Emissions as

Determined by Secondary Zone Fuel-Air Ratio

84

24

22

20

18

16

14

12

IO

8

6

4

2

0

0 0% RH; no f lag-T,=244 K (f/alp =.07 o 50%RH;vt. f lag-To=278K (f/a), =.03,.0275,.025 0 lOO%RH; hor. f lag-To=322K (f/a)d = .02 Ift. shade

.Ol5 no shade

.Ol rt. shade

i \.

a

4l 0

Q- a--

Q- %

II-- 0

o- +- I 43 m-

_;IArn. 1 I I I I I I I

600 800 1000 I200 1400 T4 -K

Figure 49. Carbon Monoxide Emissions for a Primary Zone Fuel-Air Ratio of .07

85

24

22

4

2

0

0% RH; no f lag-To=244 K (f/a),=.08

50%RH; vt. f I a g-T, q 278 K (f/a&=.035, .030, -025

IOO%RH; her f I a g-To= 322 K (f/a)d=.O20 If t. shade .Ol5 btm. shade .OlO r t. shade

600 800 1000 1200 1400 T4 - K

Figure 50. Carbon Monoxide Emissions for a Primary Zone Fuel-Air Ratio of .08

86

24

22

20

‘8

‘6

I4

I2

‘0

8

6

4

2

0

C o 0% RH; no flag-T,=244 K (f/a$,=.Og q 50%RH ; v t. f la g-To=289K (f/a),= .035, .03,.25 0 IOO%RH; hoc flag-To=322K (f/a),=.02 rt.shade

.Ol5 no shade

.Ol lft. shade

0

cl-

@-

I I

8bO I I I I I I

600 1000 ‘200 I400

T4-K Figure 51. Carbon Monoxide Emissions for a Primary Zone

Fuel-Air Ratio of .09

87

24

22

20

18

I6

I4

I2

IO

8

6

4

2

0

0 0% RH ; no flag -To = 244 K q 50%RH; vt. flag-To=289 K

(f/a+O6; (f/a)pz=.03 (f&=.02; .Ot5

0 IOO%RH; her. f lag-To=322K (f/a)d=.OlO no shade .007 btm. shade

0 0

I I I I I I I I I

600 800 1000 1200 14.00 T4 - K

Figure 52. Carbon Monoxide Emissions for a Primary Zone Fuel-Air Ratio of .06

88

24

22

20

I8

I6

I4

I2

IO

8

6

4

2

0

- 0 0% RH; no flag-To=244 K (f/a), (f/a), (f/ddshade 0 50%RH; vt. flag -To=289 K

IOO%RH hor. f lag -To 322 K (f/a)pl=.06

- 0 ; = (f/a)p;O3 ,015 a07 top

.08 .03 -01 If t.

.09 .035 -015 btm.

I I I I I I II

600 800 1000 1200 1400

T4-K

Figure 53. Carbon Monoxide Emissions for Different Primary Zones

89

DIAMETER, microns

Figure 54. Fuel Drop Fvaporation

90

1000

800

8 c 600 8

400

200

0

IO00

400

200

0

Ca Iculated: shade: Ift. btm. rl.

0% RH ; no f la g-To =244 K (f/a)p=.07 .08 .09

50%RH; vt. flag-To=289 K (f/a)s=.025 .030 .035

100% RH;hor: f lag-T,=322 K (f/a)d=.OOS .OlO .Ol5

8

Measured:

Ift. shade: (fh)d=.OO7

btm. shade: (f/a’d=.Ol I

rt. shade: (f/a)d=.015

600 800 1000 1200 1400 T4-K

Figure 55. Comparison of Calculated and Measured Carbon Monoxide Emissions

91

2.5

2.0

LJ 1.5

1.0

0.5

2.5

2.0

Meosured: (f/Q Id =.007

Calculated : ( f/a )p, (f/ds

=.06; (t/a)p,~ .03 = .015; (f/a), = .007

240 260 280 300 320

Figure 56. Ambient Temperature and Humidity Correction Factor for a Dilution Zone Fuel-Air Ratio of .007

92

2.5

2.0

0.5

2.5

2.0

1.0

0.5

-

I-

Measured: (f/a)d=.Ol I

lNTERfWLATE0

-----\

.

Calculated : (f/a),= .08 (f/cl ) s= .03 (f/a )e’= .o 1

D

240 260 280 300 320

To-K Figure 57. Ambient Temperature and Humidity Correction

Factor for a Dilution Zone Fuel-Air Ratio of .Oll

93

2.5

2.0

0.5

2.5

2.0

Measured: (fqj = .015

Calculated: lf/ajp = .09 (f/Q) (f/a);

= .Q3 = .Ql5

I I I I I I I I I I

240 260 280 300 320

To-K Figure 58. Ambient Temperature and Humidity Correction

Factor for a Dilution Zone Fuel-Air Ratio of .015

94

2.5

2.0

I.0

0.5

2.5

2.0

8 1.5 & 1.0

0.5

P.R.=4, RH=O%

Lean Primafy Zone

I I I I I I I I I I

240

Figure 59.

280 300

To - K Ambient Temperature and Humidity Correction Factor, T-56

95

137.18 +

SMD Indicated

0 0.0 RH. NO FLRG TO=244 q 0.5 RH. VT. FLAG TO=289 0 1 .O RH. HOR.FLRG TO=322

yCi;TOM SHADED 1 ;;f$-;DO; SHADED

OPEN - F/R::015

PR= 2 II3 sCONST.

117.17 e w132.92

126.68 % 124.03

P 123.26 e- 100.06

e- 129.58

B- 97.36 88.22e

92.32 o- 82.53

e- 94.83

73.730 80.82 77.26 A$ 77.41

77.59

o- 79.42

I I 1 1-l .oo 55.00 65 -00 75.00 85 -00 85 .DO

T4 - K m10'

Figure 60. Hydrocarbon Emissions, SMD Effect for a Constant Compressor Discharge Mach Number at a Pressure Ratio of Two

96

0 0.0 RH. NO FLRG TO=244 00.5 RH, VT. FLAG TO=289 0 1 .O RH. HOR.FLAG TO=322

;;;TOM SHRDED - F/A=.007 SHADED - F/R=.011

OPEN - F/A=.015

PRI 3 M3 XONST =

96.72 e +ogs56

102.04 Et- 107.36

102.18

e- 105.01 77.55

78.0 ' e 77.35

O-7;5y08e139.35 .

e- 79-07

145.1 142.58 144.4 ?a?-

144.850145.35 I I I I

i.00 65 .oo 75 l oo 85.00 95 000 T4 - K d0'

--co5 .oo

Figure 61. Hydrocarbon Emissions, SMD Effect for a Constant Compressor Discharge Mach Number at a Pressure Ratio of Three

97

92.05 A

I 0 0.0 RH, NO FLRG TO=244 Cl 0.5 RH m VT. FLAG TO=289 0 1 .O RH. HOR.FLAG TO=322

BOTTOM &i;fW; - F/A=.007 TOP - F/R=.011 OPEN - F/R=.015

~ PR= 4 M3 SXINST l

I

0 83.12

r;+ 91.01 86.87 e- 71.90

e 156.65 6.55 6.87

164.85 a- 88.90 139.54 0 R-

141.06 147.96 0172.21 145.49 163.46

167.41 I I I 5.00 65.00 75.00 85.00 95 -00 105 .oo

T4 - K d0'

Figure 62. Hydrocarbon Emissions, SMD Effect for a Constant Compressor Discharge Mach Number at a Pressure Ratio of Four

98

SMD Indicated

1.00

0 0.0 RH. NO FLftG TO=244 q 0.5 RH. VT. FLAG TO=289 0 1 .O RH. HOR.FLAG TO=322

$TOM W&; - F/A=.007

OPEN - F/R=.011 - F/A=.015

PR= 5 M3 =CONST.

A 82.48

974.73

174.31 e

n- 80.22 o- 161.71

77.40 77.58

‘---I- -~ 70.00

77.62 *

168.08 g-79.03 7v

167.04 -~ I --_- b'u~~ 181.70 80 .OO 90 -00 100 .oo lin.nn

T4 - K MO’ Figure 63. Hydrocarbon Emissions, SMD Effect for a

Constant Compressor Discharge Mach Number at a Pressure Ratio of Five

99

Combustion Efficiency Indicated

0 0.0 RH. NO FLAG TO=244

PR= 2 M3 =CONST l

o- 97.31

-95.14

96.19

96.23 e 91.70

93.45

6 89.34 &I 93.81

90.31 0 4 93.38

89.95 92.70 Q

87.94 la- 87.77 5 92.14

o- 83.46 u- 89.36

6 96.16 0 95.34

R- 95.61

o- 93.49

I-1 ----------.-.. _- __-_ _ ~ 65.00 75 .oo ak moo 95 .oo

T4 - K d0'

Figure 64. Oxides of Nitrogen, Combustion Efficiency for a Constant Compressor Discharge Mach Number at a Pressure Ratio of Two

100

Combustion Efficiency Indicated

0 0.0 RH. NO FLAG TO=244 El 0.5 RH. VT. FLRG TO=289 0 1 .O RH. HOR.FLAG TO=322

BOTTOM SHRDED - F/R-.007 TOP SHRDED - F/R=.011 OPEN - F/R= .015

PR= 3 M3 =CONST.

o-99.22

5 98.33

6 98.69

e- 96.02 .

(N x 0 6 97.26 98.64

z I 6 98.58

GE 97.08 . N Q

l+ 94.29

97.08 0 97.16

93.84

z 93.65 e95.00 .

A D98.51

8 90.61 IT- 96.98

E93.65

95.12 o- 97.42

I

&go,23

65x0 I e, I

75 .oo 65.00 951 105.00 T4 - K x10’

Figure 65. Oxides of Nitrogen, Combustion Efficiency for a Constant Compressor Discharge Mach Number at a Pressure Ratio of Three

101

Combustion Efficiency Indicated

I-- 0 0.0 RH, NO FLRG TO=244 U0.5 RH. VT. FLRG TO=289 0 1 .O RH. HOR.FLRG TO=322

BOTTOM SHADED - F/R=.007 TOP SHADED - F/R=.011 OPEN - F/R= -015

PR= 4 M3 =CONST.

o- 99.79

e- 99.52

99.69

e98.57

6 99.19 tl 99.65

0 97.66

b 97.50

0 97.02

6 99.60

98.46

6 98.96 0 98.82

cl- 99.46

e 97.43 @ 95.67 m- 98.77

E 96.94

97.16 u 98.54 93.94

I I -7------l .oo 65 .OO 75 .oo 85 .OO 95.00 1.05 .oo

T4 - K 40' Figure 66. Oxides of Nitrogen, Combustion Efficiency

for a Constant Compressor Discharge Mach Number at a Pressure Ratio of Four

102

Combustion Efficiency Indicated

0 0.0 RH. NO FLAG TO=244 q 0.5 RH. VT. FLRG TO=289 0 1 .O RH. HOR.FLRG TO=322

;;;TOM SHADED - F/R=.007 SHADED - F/R=.011

OPEN - F/R=.015

PR= 5 M3 =CONST.

e- 99.75

.oo-

o- 99.85

e- 99.81

99.80

99.60 tl 99.77

99.29 I&! 99.50 99.78

6

P 99.12 4 99.51 0 99.43 99.38

Q 99.00 8 99.48 III-- 99.70

99.46 0 97.63 IT- 99.50

97.86 Id- 98.72

o- 96.89

e- o- 99.25

98.49

-- T 1 I I 70 -00 80 -00 90.00 100.00 110.00

T4 - K m10'

Figure 67. Oxides of Nitrogen, Combustion Efficiency for a Constant Compressor Discharge Mach Number at a Pressure ratio of Five

103

1. Report No.

NASA CR-3355 4. Title and Subtitle

2. Gove%nent Accession No.

.- --- -- -.

3. Recipient’s Catalog No.

EFFECT OF AMBIENT CONDITIONS ON THE EMISSIONS FROM A GAS TURBINE COMBUSTOR

7. Author(s) I

8 Performing Organization Report No.

C. W. Kauffman

9. Performing Organization Name and Address

University of Cincinnati 11, Contract or Grant No.

Cincinnati, Ohio 45219 NSG- 3045

12. Sponsoring Agency Name and Address

National Aeronautics and Space Administration Washington, D. C. 20546

13. Type of Report and Period Covered

Contractor Report 14. Sponsoring Agency Code

I _;~- __- 15. Supplementary Notes

Final report. Project Managers, Robert E. Jones, Aerothermodynamics and Fuels Division, Lewis Research Center, and David B. Ercegovic, U.S. Army Research and Technology Laboratories (AVRADCOM), Propulsion Laboratory, 44135.

Lewis Research Center, Cleveland, Ohio

16. Abstract

An experimental and analytical investigation was initiated to determine, in a systematic way and under controlled conditions, the effect of variations in the ambient conditions of pressure, tem- perature, and relative humidity upon the emissions of a gas turbine combustor. A single com- bustor can from a Pratt and Whitney JT8D-17 engine was run at parametric inlet conditions bracketing the actual engine idle conditions. Data were correlated to determine the functional relationships between the emissions and ambient conditions. Mathematical modelling was used to determine the mechanism for the carbon monoxide and hydrocarbon emissions. Carbon monoxide emissions were modelled using finite rate chemical kinetics in a plug flow scheme. Hydrocarbon emissions were modelled by a vaporization scheme throughout the combustor.

7. Key Words (Suggested by Author(s) J

Emissions Combustors Modelling

16. Distribution Statement

Unclassified - unlimited STAR Category 07

9. Security Classif. (of this report)

Unclassified 20. Security Classif. (of this page)

*For sale by the National Technical Information Service, Spinefield, Virginia 22161 NASA-Langley, 1960