https://ntrs.nasa.gov/search.jsp?

R=19780019187 2019-07-31T15:41:16+00:00Z

NASA
TP
1260
c.1
NASA AVRADC&
Technical Paper 1260 Technical Report 78-22(PL)

Reverse-Flow Combustor for

p
Small Gas Turbines With
Pressure-Atomizing Fuel Injectors

Carl T. Norgren, Edward J. Mularz, ~

and Stephen M. Riddlebaugh

JULY 1978

NASA
 0234349
NASA AVRADCOM
Technical Paper 1260 Technical Report 78-22(PL)

Reverse-Flow Combustor for

Small Gas Turbines With
Pressure-Atom izing Fuel Inj ectors

Carl T. Norgren
Lewis Research Ceiiter, Clevelaiid, Ohio
Edward J. Mularz
Propzrlsioii Laboratory, A V R A D C O M Research mid Techriology Laboratories
Lewis Research Ceiiter, Clevelaiid, Ohio
Stephen M. Riddlebaugh
Lewis Research Ceiiter, Cleveland, Ohio

NASA
National Aeronautics
and Space Administration

Scientific and Technical

Information Office

1978
 REVERSE-FLOW COMBUSTOR FOR SMALL GAS TURBINES
WITH PRES S URE-ATOMIZ ING FUEL INJECTORS
by C a r l T. Norgren, Edward J. Mularz,* a n d Stephen M. Riddlebaugh
Lewis Research Center

SUMMARY

A reverse-flow combustor suitable for a s m a l l gas turbine (2 to 3 kg/s m a s s flow)

was used t o evaluate the effect of pressure-atomizing fuel injectors on combustor per-
formance. In these tests a n experimental combustor was designed to operate with 18 s i m -
plex pressure-atomizing fuel injectors at sea-level takeoff conditions. To improve per-
formance at low-power conditions, the fuel manifolding was modified s o that only every
other injector was operational. Combustor performance, emissions, and liner temper-
a t u r e are compared over a range of p r e s s u r e and inlet-air conditions corresponding to
simulated idle, cruise, and takeoff typical of a 16 to 1 p r e s s u r e ratio turbine engine.
At simulated combustor inlet sea- level takeoff conditions with a n overall fuel-air
ratio of 0.014 the combustor with 18 fuel injectors operated with a combustion efficiency
of 100 percent, a total p r e s s u r e loss of 1 . 7 percent, a pattern factor of 0.21, and emis-
sion levels of unburned hydrocarbons, carbon monoxide, and oxides of nitrogen of 0.1,
1.5, and 16 grams per kilogram of fuel, respectively, with a smoke number of 22. It
was not possible t o operate this combustor with 18 injectors at idle because of blowout
caused by low fuel p r e s s u r e and subsequent fuel s p r a y deterioration. By reducing the
fuel injector density it was possible t o operate at idle. At idle the combustor with nine
fuel injectors operated with a combustion efficiency of 92.4 percent, a total p r e s s u r e
loss of 1 . 3 percent, a pattern factor of 1.12, and emission levels of unburned hydro-
carbons, carbon monoxide, and oxides of nitrogen of 62.9, 96.9, and 3.9 g r a m s per
kilogram of fuel, respectively, with a smoke number of 24.

INTRODUCTION

As p a r t of a continuing effort at the Lewis Research Center t o improve performance,

emissions, and- reliability of turbine machinery, a n investigation has been initiated to
* Propulsion Laboratory, U. S. Army R&T Laboratories (AVRADCOM)).
I --
I I1

provide design criteria for small gas turbine combustors. The performance and pollu-
tant emission levels a r e documented over a range of simulated flight conditions for a
reverse-flow combustor configuration using simplex pressure-atomizing fuel injectors
in which the effect of reducing the number of fuel injectors to improve idle performance
has been investigated.
Small turbine machinery, whether used for primary propulsion or auxiliary power
application, is inherently l e s s efficient than its larger sized counterpart due to the s m a l l
s i z e of the components, manufacturing tolerances, and material limitations (ref. 1).
An improvement i n cycle efficiency can be obtained, however, with increased compres-
sion ratio, higher combustor inlet-air temperature, and increased turbine inlet tem-
perature. Techniques used to improve performance characteristics of large combustors
are not necessarily applicable to small combustors because of the problems which arise
due to scaling. Problem a r e a s of particular interest i n small combustor technology
include such items as combustion stability, liner cooling, and temperature distribution.
Combustion stability is affected by many factors including method of fuel introduction,
fuel and air distribution, and wall quenching. Liner cooling is of particular concern
because of the high combustor surface to volume ratio and the trends toward high tur-
bine inlet temperature t o improve overall cycle efficiency. A uniform outlet tempera-
t u r e distribution is sensitive t o mixing, upstream passage blockages, and perturbations
in the flow.
While many types of combustor configurations a r e possible for use in small turbine
machinery, the reserve-flow configuration offers several advantages (ref. 2). As a
technique to improve performance and reduce manufacturing difficulty, small turbine
machinery usually incorporates a final centrifugal stage in the compressor. A f i n a l
centrifugal stage coupled with a reverse-flow combustor permits the use of a radial
diffuser, which is highly efficient with respect t o reducing pressure losses in the dif-
fuser and improving flow distribution into the combustor. The reverse flow combustor
provides a larger combustion volume than would otherwise be obtained with a straight
through flow combustion chamber; thus, a potential gain in performance can be realized.
In addition, engine packaging is favorably affected by permitting a shortening of the
rotating shaft and by placing the fuel injectors in a readily accessible location.
Fuel distribution remains a problem in the reverse-flow combustor because of the
relatively large number of fuel injectors required t o effectively distribute the fuel at the
large combustor diameter t o obtain the uniform outlet temperature distribution required
for takeoff. Simplex pressure-atomizing fuel injectors sized for takeoff cannot produce
satisfactory spray characteristics at idle fuel flows. On the other hand, if the injectors
a r e sized for idle fuel flows, the physical size of the fuel passages within each injector
would be so small that excessive pressure losses would occur at higher flow rates and
reliability would be adversely affected because of the increased susceptibility to clogging.

2
A technique which has been used i n swirl-can combustors t o improve efficiency at idle
is t o provide locally idealized burning zones (ref. 3). A similar technique to improve
the fuel-air ratio in localized zones within the combustor was used i n this investigation.
In this technique a portion of the fuel injectors is rendered inoperable and the increased
fuel flow to each of the remaining injectors provides improved spray characteristics.
To investigate the design criteria envolved i n s m a l l combustor technology a reverse-
flow combustor configuration was selected. The design of the reference combustor was
based in as far as feasible on c u r r e n t technology. An investigation of varying fuel in-
jector density was incorporated into the first phase of the program. The p r e s s u r e
atomizing injectors were sized for sea-level takeoff conditions, and it was anticipated
that idle performance would deteriorate. Fuel injector density was varied by operating
the combustor with 9 o r 18 evenly spaced simplex pressure-atomizing fuel injectors.
The effect on combustion efficiency, combustor total pressure loss, outlet-temperature
profiles, and liner wall temperature was investigated for simulated idle, two cruise
conditions, and takeoff for a 16 t o 1 compression r a t i o gas turbine engine. A para-
metric variation of combustor reference velocity was a l s o included. Exhaust emission
levels of unburned hydrocarbons, carbon monoxide, nitrogen oxides, and smoke number
were obtained. Although reducing the number of fuel injectors would be required only
at idle, there is a possibility that this technique might improve performance at condi-
tions other than idle; therefore, the effect of varying the fuel injector density was in-
vestigated at all test conditions.

APPARATUS

Test Facility

The test combustor was mounted i n the closed-duct facility shown schematically i n
figure 1. The laboratory air supply can maintain airflow rates up to 15 kilograms per
second at pressure levels up to 3000 kilopascals. Tests w e r e conducted up to p r e s s u r e
levels of 1600 kilopascals. For these t e s t s combustion air drawn from the laboratory
high-pressure supply was indirectly heated t o 717 K i n a counterflow tube heat exchanger.
The temperature of the air flowing out of the heat exchanger was automatically controlled
by mixing the heated air with varying amounts of cold bypassed air. Airflow through the
heat exchanger and bypass flow system and the total p r e s s u r e of the combustor inlet
airflow were regulated by remotely controlled valves.

3
Il ! ! ! 111 I l11l1l11llIII

Test IEardware

A c r o s s section of the reverse-flow combustor used i n this investigation is shown

in figure 2(a). An isometric sketch of the reverse-flow combustor is shown i n fig-
u r e 2(b). The combustor is a n experimental NASA design with a maximum diameter of
38.5 centimeters. The design s t r e s s e s versatility so that modification or replacement
of the swirlers, faceplate, liner, and turning sections can be accomplished. The design
liner p r e s s u r e loss is 1 . 5 percent and the diffuser dump loss is 0.24 percent. A sym-
metrical hole pattern based on 36 circumferential locations is used s o that 9, 12, o r
18 fuel injectors can be, staged and still maintain a symmetrical air entry pattern (in
this investigation the combustor is operated with 9 or 18 injectors, see fig. 2(c)). The
airflow distribution and hole s i z e s a r e shown in table I. Photographs of the reverse-
flow combustor and housing are shown in figure 3. In figure 3(a) an aft view of the
combustor is shown before assembly. In figure 3(b) the combustor is shown partly
assembled, and in figure 3(c) it is shown completely assembled.
In the first phase of this study, pressure-atomizing simplex fuel injectors are used
with flow r e s t r i c t o r s ahead of each injector t o aid i n obtaining an even distribution of
fuel. The injector s p r a y angle was ' 5*7 and the orifice was sized to provide a flow
of 0.0032 kilogram p e r second (25 lb/hr) for a p r e s s u r e drop a c r o s s the orifice of
690 kilopascals (100 psid).

Instrumentation

The combustor instrumentation stations a r e shown i n figure 4. Five total p r e s s u r e

probes, two static pressure taps, and five Chromel-Alumel thermocouples a r e located
at station 2 to measure the inlet temperature and pressure. At station 3 a s e r i e s of
18 total p r e s s u r e probes a r e installed t o determine the inlet-air profile and t o deter-
mine the extent of any flow disturbance behind the s t r u t s which support the centerbody
diffuser. At station 4 six Pitot-static probes a r e positioned in the cold-air passages
between the combustor liner and combustor housing t o determine passage velocity and
distribution. Four gas sample probes, evenly spaced on the circumference, a r e a l s o
located at station 5.

PROCEDURE

Test Conditions

The experimental reverse-flow combustor was operated at test conditions based on

a gas turbine engine cycle with a compressor pressure r a t i o of 16 to 1. A tabulation of
4
the test conditions simulated i n this study are shown i n table II.
Data were obtained at combustor inlet conditions simulating sea-level takeoff,
cruise, and idle. Data were obtained over a range of fuel-air ratios from about 0.009
to 0.016. However, because of thermocouple limitations, the overall fuel-air ratio was
limited to approximately 0.014 at sea-level takeoff. At the idle condition the fuel-air
ratio was 0.01. The combustor was operated with a parametric variation of reference
velocity at sea level and cruise (24 and 30 m/s in addition t o the reference velocity of
18 m/s). The reference velocity quoted is based on the assumption of unidirectional
total m a s s flow and maximum cross-sectional area of the housing prior to the r e v e r s e
turn as shown i n figure 2(a). The combustor was also operated at simulated reduced
power at a constant fuel-air ratio of 0.014. For the reduced power conditions a pres-
sure level lower than cruise was selected, and the corresponding inlet temperature was
calculated using a compressor efficiency of 80 percent. Also presented in table II a r e
the simulated compressor pressure ratios. These ratios as presented are referenced
to sea-level pressure.
The combustor was operated with two fuel staging configurations. Combustor
models A - 1 and A-2 had 18 injectors each, but i n model A-2 every other injector w a s
made inoperative by closing off its fuel lines (see fig. 2(c)). For check purposes to
determine the effect of combustor symmetry on performance, the nine even nozzles and
the nine odd nozzles were activated in turn, and the combustor performance was com-
pared. Since no change in performance was expected or experienced, only one s e t of
data is presented for combustor model A-2. The test program was conducted using
Jet A fuel.

E mission Measurements

Exhaust gas samples were obtained according to the procedures recommended in

references 4 and 5. Exhaust gases were withdrawn through four air-cooled stationary
probes mounted approximately in the stator plane and in the center of the exhaust duct
(see fig. 4). Concentrations of oxides of nitrogen, carbon monoxide, and unburned
hydrocarbons w e r e determined with the gas analysis described in reference 6. The gas
sample temperature was held at approximately 423 K in the electrically heated sampling
line. Most of the gas sample entered the analyzer oven, while the excess flow was
bypassed t o the exhaust system. To prevent fuel accumulation in the sample line, a
nitrogen purge was used just before and during combustor ignition.
After passing through the analyzer oven, the gas sample was divided into three
parts, and each part was analyzed. Concentrations of oxides of nitrogen, carbon mon-
oxide and carbon dioxide, and hydrocarbons were measured by the chemiluminescence,

5
I I1 Ill I I 1 1 Illl1l1l1l1l11ll11l Ill I1 I1 I I

nondis per s e d- infrar ed, and flame -ionization methods , respectively .

Gas samples used to determine oxides of nitrogen and carbon monoxide were passed
through a refrigerated dryer and analyzed on a d r y basis. Readings for oxides of
nitrogen and carbon monoxide were corrected so that they could be reported on a wet
basis, as were those for unburned hydrocarbons.
Fuel-air ratios calculated from a carbon balance agreed to within 10 percent with
values obtained from fuel-flow and airflow measurements. The combustion efficiency
data presented in this report were based on stoichiometry determined by gas analysis.

RESULTS

The following data were obtained in the reverse-flow combustor to investigate the
effects of fuel staging with pressure atomizing simplex fuel injectors. Data were ob-
tained for simulated inlet conditions typical of operating a 16 t o 1 pressure ratio turbine
engine at the test conditions tabulated i n table I. The outlet temperature level was
limited to approximately 1350 K because of instrumentation constraints. The combustor
was operated with 18 and 9 evenly spaced simplex fuel injectors, designated as com-
bustor models A-1 and A-2, respectively, using Jet A fuel. Performance and emission
data a r e presented in figures 5 to 14 for simulated inlet flight conditions. Idle data are
tabulated in table III. The experimental performance and emissions data are tabulated
in table IV.

Performance

Combustion efficiency. - Combustion efficiency data are shown in figure 5. In

figure 5(a) the efficiency is shown for a range of fuel-air ratios. The combustion effi-
ciency is essentially independent of fuel-air ratio over the range of fuel flows investi-
gated for all simulated flight conditions with a combustion efficiency level greater than
99 percent except for the high altitude cruise which dropped somewhat at low fuel-air
ratios.
The effect of off-design operation at lower compressor pressure ratio on combustion
efficiency is shown in figure 5(b). The reference velocity and fuel-air ratio were held
constant at 5.5 meters per second (18 ft/sec) and 0.014, respectively, as indicated in
table 11. A s shown in figure 5(b), the combustion efficiency fell off sharply for pressure
ratios below 8 . 5 t o 1 with model A-1 (18 fuel injector configuration); blowout occurred
before the idle pressure level could be achieved. With model A-2 (9 fuel injector con-
figuration) the combustion efficiency had fallen off to approximately 94 percent at a

6
 pressure ratio of 4.1 to 1. Note the test condition in figure 5(b) at a 4.1 t o 1 pressure
ratio corresponds approximately to idle; differences occur in fuel-air r a t i o (0.014as
compared to 0.01 at idle) plus nominal differences in m a s s flow (see table II). A t the
idle conditions the combustion efficiency for model A-2 was 92.4 percent (see table III).
At these lower pressure levels the combustion efficiency is sensitive to further pressure
reductions as noted by the sharp dropoff in efficiency (fig. 5(b)).
The reverse-flow combustor was a l s o operated with increased airflow loading (in-
creased reference velocity). A parametric variation of increased mass flow rates
corresponding to increases in reference velocity from 5 . 5 t o 7.32 or 9.14 meters per
second had no appreciable effect on combustion efficiency (see table IV).
P r e s s u r e loss. - The reverse-flow combustor pressure loss data a r e shown in
figure 6. The total p r e s s u r e l o s s for the design mass loading at a reference velocity of
5.5 meters per second (diffuser inlet Mach number of 0.054) was 1.2 percent. A 67-
percent increase i n reference velocity increased the total pressure loss by approxi-
mately 0.4 percent over the range of temperature r i s e levels investigated.
Outlet temperature distribution. - The data obtained for the pattern factor are
shown in figure 7. The pattern factor is of the order of 0.21 t o 0.24 with model A- 1
and approximately 0 . 6 with model A-2 as shown in figure 7(a) over the range of fuel-air
ratios investigated at a reference velocity of 5 . 5 meters per second. The effect of
increased reference velocity is shown in figure 7(b). At a reference velocity of
9.14meters per second the pattern factor was degraded with model A-1 to approxi-
mately 0.3; with model A-2, however, the pattern factor was improved to approximately
0.46. These changes were probably due t o the higher mass loading of the combustor
at the higher reference velocity. In model A - 1 the dilution zone was designed for a
lower reference velocity. An increase in reference velocity caused excessive penetra-
tion, which resulted in a poorer temperature distribution. With model A-2 a poorer
fuel distribution existed in the primary and the increased mixing intensity would aid in
dispersing hot zones.
Typical radial temperature profile data a r e shown in figure 8 for simulated takeoff
conditions. Included in figure 8 is a n ideal profile reproduced from one which is con-
sidered desirable for larger turbine machinery (ref. 6) and would a l s o be applicable
for small engines considering fatigue, creep, and erosion of the turbine blade. Both
combustor models A-1 and A-2 produced s i m i l a r outlet profiles that were slightly
cooler than the required profile at the hub.
Liner temperature. - The combustor liner temperature pattern was observed with
thermal paint. From this observation the thermocouple locations were selected on the
basis of the highest liner temperature level indicated for a particular liner location.
Six locations were selected t o observe the liner temperature - the primary combustion
zone, the outer turn, and the inner turn; specific locations are indicated in figure 2.

I
In all cases the inner turn exhibited the highest wall temperature. Typical liner tem-
peratures observed at this location are shown i n figure 9. As the t e s t conditions in-
creased i n severity or the fuel-air ratio level increased the peak liner temperature in-
creased i n a uniform manner. With model A-1 for the sea-level takeoff inlet conditions
the maximum liner temperature was 908 K (1175' F) at a n overall fuel-air ratio of
0.014. Similar trends were obtained with combustor model A-2 as indicated i n figure 9.

Emissions

- . _-
Unburned hydrocarbons. - The emission index data for unburned hydrocarbons are
shown in figure 10. In order to compare the data for model A-2 with model A-1 at a
fuel-air r a t i o of 0.014 it was necessary t o extrapolate the A-2 data obtained at lower
fuel-air ratios. In figure lO(a)the effect of compressor pressure ratio is shown for
combustor models A- 1 and A-2. As the p r e s s u r e level was reduced with model A- 1
the hydrocarbon emission index increased rapidly below a compression ratio of 8.5 to
1. With combustor model A-2 the increase i n hydrocarbon emission occurred at a
lower compressor pressure ratio (approximately 6 t o 1) and the rate of increase was
significantly reduced. At the sea-level takeoff condition, combustor model A- 1 oper-
ated with an unburned hydrocarbon emission index of 0.1 gram per kilogram of fuel. At
the idle condition with combustor model A-2 the hydrocarbon emission index was
62.9 g r a m s per kilogram of fuel (see table In). As previously noted, model A-1 could
not be operated at idle.
The effect of increased reference velocity is shown in figure 10(b). No appreciable
hydrocarbon emission was observed at the cruise or sea-level takeoff condition. The
maximum hydrocarbon emission index of 0.63 gram per kilogram of fuel with model A-1
was obtained at the high altitude cruise condition.
Carbon monoxide. - The emission index data for carbon monoxide are shown in
figure 11. In figure ll(a) the effect of compressor p r e s s u r e ratio is shown for combus-
tor models A-1 and A-2. A s the pressure and temperature levels were reduced with
models A-1 and A-2 the carbon monoxide level increased. The shapes of the two curves
a r e similar; however, the carbon monoxide level with model A-2 was significantly lower
than it was with model A-1 (approximately 1 g/kg of fuel f o r model A-2 as compared to
15 g/kg of fuel with model A-1 at a compressor p r e s s u r e ratio of 10 to 1). At the simu-
lated sea-level takeoff condition model A-1 produced an emission index value of 2 g r a m s
p e r kilogram of fuel. At the idle condition with combustor model A-2 the carbon mon-
oxide emission index was 96.9 g r a m s per kilograms of fuel (see table III).
The effect of increased reference velocity is shown in figure l l ( b ) . As the com-
pressor pressure increased (corresponding t o a decrease in altitude) or as the reference

8
 velocity increased, the carbon monoxide emission index decreased for model A- 1. The
extrapolated data for model A-2 indicated a decrease in the carbon monoxide level with
increased pressure and, in general, a slight increase in the carbon monoxide level at
the higher reference velocity.
Oxides of nitrogen. - The emission index data for the oxides of nitrogen are shown
in figure 12. In figure 12(a)the effect of compressor pressure ratio is shown for com-
bustor models A-1 and A-2. As the pressure and temperature levels were reduced the
oxides of nitrogen decreased. Model A-1 exhibited a lower level of the oxides of nitro-
gen than did model A-2. The lower level with model A-1 may i n part be attributed to
the lower combustion efficiency and reduced inlet temperatures at reduced pressure
ratios; however, at a condition where the combustion efficiency was essentially
100 percent (compressor pressure ratio of 1 0 to 1 - corresponding t o high altitude
cruise), the emission index was approximately 10 grams per kilogram of fuel for model
A-1 and 1 2 grams per kilogram of fuel for model A-2. At sea-level takeoff model A-1
produced an emission index value of 16 g r a m s per kilogram of fuel. At the idle condi-
tion with combustor model A-2 the oxides of nitrogen emission index was 3 . 9 grams
per kilogram of fuel (see table HI).
In figure 12(b) the effect of increased reference velocity on the emission of oxides
of nitrogen is shown. In general, at the higher reference velocity the oxides of nitrogen
decreased. The extrapolated emission index level for model A-2 was slightly higher
than that for model A-1 at the cruise condition. A t sea-level takeoff a comparison
between models A-1 and A-2 was not obtained because of the sensitivity of the oxides of
nitrogen emission with fuel-air ratio. It was considered that the extrapolation required
for model A-2 would be misleading.
-Smoke
__ number. - The smoke number data are shown in figure 13. The combustor
model A-1 produced smoke levels of the order of 9 to 25 over the range of test condi-
tions. Combustor model A-2 produced smoke levels from 24 to 50 over a similar range
of test conditions. A t sea-level takeoff model A - 1 produced a smoke number of 22. At
the idle condition with combustor model A-2 the smoke number w a s 24 (see table III).
It would be expected that the increase i n smoke observed with model A-2 would be due
t o the locally rich zones s e t up by fuel injector spacing.

DIS CUSSION

Performance

Reducing the number of fuel injectors was investigated t o determine the improve-
ment i n combustor performance which could be obtained at low power conditions by

I
 improving the fuel injector spray characteristics. As previously discussed, it was not
possible to operate model A-1 at idle conditions because of blowout, which was attri-
buted to poor s p r a y characteristics. The pressure drop data a c r o s s the injectors used
in combustor models A-1 and A-2 are shown in figure 14. The idle total fuel flow was
of the order of 44 kilograms per hour (97 lb/hr). The corresponding pressure drops
were approximately 27 kilopascals (4psid) for model A-1 and 122 kilopascals (17.9 psid)
for model A-2. The injector pressure drop for model A-1 is included for comparison
even though sustained combustion was not possible. As fuel flow requirements in-
creased, the pressure drop a c r o s s the injectors increased at a considerably faster r a t e
for model A-2 than for model A-1. As shown in figure 5(b), the combustion efficiency
for model A-2 approached 100 percent at a much faster r a t e than did model A-1 as the
compressor pressure increased.
Based on a combustion efficiency viewpoint, it is apparent that performance began
t o deteriorate at low fuel injector pressure. However, it could be improved by reducing
the number of injectors and thereby increasing the pressure differential for a given fuel
flow. The extra injectors were not removed for the A-2 configuration nor was the com-
bustor geometry altered. Fuel flow to the extra injectors was cut off, but the air flow
was not. However, combustor performance was not adversely affected and stability was
improved. It is probable that fuel injector programming of the type investigated in this
study could be incorporated into a combustion system to improve low-power combustion
efficiency.
There was no apparent effect of varying the number of fuel injectors on total pres-
sure loss with either model A-1 o r model A-2.
The outlet temperature profile obtained with model A-1 was considered satisfactory.
The outlet temperature profile with model A-2 deteriorated; however, even with an idle
pattern factor of 1.12, a maximum temperature of only 1300 K (1940' F)would be ex-
perienced at the turbine stator plane because of the lower temperature levels associated
with reduced power. The value of 0 . 6 for the pattern factor at cruise for model A-2
would be unacceptable because of the high turbine inlet requirements (turbine inlet tem-
perature, 1565 K (2360' F)). The high pattern factor points out that in this configura-
tion (model A-2)nine fuel injectors could not provide the outlet temperature distribution
required for flight conditions. The radial profile for model A-2 was acceptable, how-
ever.

Emissions

A t low-power conditions the emission levels of unburned hydrocarbons and carbon

monoxide a r e usually of most concern. From the data obtained with model A-1 it is

10
 apparent that the unburned hydrocarbons level is unusually high and accounts for most
of the inefficiency as compared t o the carbon monoxide level. This would be consistent
with the deterioration of the fuel s p r a y at a low injector p r e s s u r e level. With the model
A-2 combustor the increase in the unburned hydrocarbon level is reduced as compared
to the carbon monoxide level; this condition indicates that the carbon monoxide reaction
is quenched rather than never initiated as would be the case with unburned fuel droplets.
The fuel spray could probably be further improved with model A-2 as evidenced by the
still high unburned hydrocarbon emission index at idle (62.9 g/kg of fuel).
At the high-power conditions the oxides of nitrogen are essentially the major pollu-
tant. In this combustor configuration the oxides of nitrogen reached a peak around a n
emission index of 16 g r a m s p e r kilogram of fuel f o r model A-1. At the low-power con-
ditions a higher index level was observed with model A-2 than with model A-1. At high
reference velocities the emission index decreased as would be expected with a shorter
residence time. The oxides of nitrogen were higher with combustor model A-2 than
with model A-1. In model A-2 locally r i c h zones are set up which contribute t o higher
oxides of nitrogen levels by increasing flame temperature. At the higher fuel flows the
emission level is also high, probably due to preferential high flame temperature s u r -
rounding the r i c h core. Improving the distribution of fuel and air, as i n model A-1
(18 injectors), provides better mixing, which is accompanied by a reduction in the oxides
of nitrogen (see ref. 7).
The effect of f u e l injector density on smoke number was to increase the smoke level
with model A-2. It w a s suspected that the increase in smoke with model A-2 would in
p a r t be due to the locally rich zones. The smoke produced in the locally rich zones dur-
ing combustion could also be prematurely quenched before any appreciable burnout
would occur because of the cool zones produced by swirlers in which the f u e l injectors
were inoperative.

SUMMARY O F RESULTS

A reverse-flow combustor suitable for a s m a l l gas turbine engine was used t o eval-
uate the effects of fuel injector density on combustor performance and emissions. Data
were obtained at p r e s s u r e and inlet-air temperature levels corresponding t o simulated
idle, cruise, and takeoff conditions of a 16 t o 1 pressure ratio engine. The outlet tem-
perature was limited t o approximately 1350 K because of the instrumentation. The fol-
lowing results were obtained with 9 and 18 evenly spaced simplex pressure-atomizing
, fuel injectors using Jet A fuel:
1. It was not possible t o operate at idle with 18 fuel injectors because of blowout.
2. The combustor with 18 fuel injectors at simulated sea-level takeoff conditions

11
operated with a combustion efficiency of 100 percent, a total p r e s s u r e loss of 1 . 7 per-
cent, a pattern factor of 0.21, and emission levels of unburned hydrocarbons, carbon
monoxide, and oxides of nitrogen of 0.1, 1.5, and 16 g r a m s per kilogram of fuel, re-
spectively, with a smoke number of 22.
3. The combustor with 9 fuel injectors at idle operated with a combustion efficiency
of 92.4 percent, a total pressure loss of 1 . 3 percent, and emission levels of unburned
hydrocarbons, carbon monoxide, and oxides of nitrogen of 62.9, 96.9, and 3.9 g r a m s
per kilogram, respectively, with a smoke number of 24.

Lewis Research Center,

National Aeronautics and Space Administration,
Cleveland, Ohio, March 15, 1978,
505- 04.

REFERENCES

1. Munt, Richard; Danielson, Eugene; and Deimon, J a m e s : Aircraft Technology As-

sessment; Interim Report on the Status of the Gas Turbine Program. U. s. En-
vironmental Protection Agency, Dec. 1976.
2. Norster, E. R. ; and Demetri, E. P. : Analysis and Design of Combustors for Small
Gas Turbines for Advanced Helicopters. Rep. 1208-1, Northern Research and
Engineering Corporation, 1974.
3. Niedzwiecki, Richard W.; Trout, Arthur M. ; and Mularz, Edward: Performance of
a Swirl-Can Combustor at Idle Conditions. NASA TM X-2578, 1972.
4. Control of Air Pollution from Aircraft and Aircraft Engines - Emission Standards
and T e s t Procedures f o r Aircraft. Federal Register, vol. 38, no. 136, pt. 2,
Tues., July 17, 1973, pp. 19088-19103.
5. Procedure for the Continuous Sampling and Measurement of Gaseous Emissions From
Aircraft Turbine Engines. Aerospace Recommended Practice 1256, Oct. 1971,
SAE.
6. Ingebo, Robert D. ; and Norgren, C a r l T. : High P r e s s u r e Combustor Exhaust Emis-
sions with Improved Air- Atomizing and Conventional Pressure- Atomizing Fuel
Nozzles. NASA TN D-7154, 1973.
7. Jones, Robert E. : Advanced Technology for Reducing Aircraft Engine Pollution.
NASA TM X-68256, 1973.

12
 TABLE I. - LINER AIRFLOW DISTRIBUTION
I A T 'ercent of total
m a s s flow
Comments

11.2
Primary Primary holes 24.88

Dilution Dilution holes 30.80

Concentric around Annulus 3.17

fuel injector
Liner cooling Film cooling 13.12
Outer 180' Film cooling 13.84
Inner 180' Film cooling 3.02

TABLE II. - REVERSE-FLOW COMBUSTOR TEST CONDITIONS

:nlet pressure ieference velocity Simulated compressor Comments
pressure ratio
kPa psia m/s ft/sec
2.27 5 1014 147 686 775 5.5 18 10 to 1 High-altitude cruise
3.05 6.71 1358 197 703 805 5.5 18 13.4 to 1 Low-altitude cruise
3.63 8 1620 235 717 830 5.5 18 16 to 1 Sea-level takeoff
1.23 2.70 405 58.8 474 394 5.2 16.9 4 to 1
2.12 4.66 862 125 627 668 5.5 18 8.5 to 1
Simulated reduced
1.83
1.51
1.23
4.02
3.33
2.70
689
5 17
414
100
75
60
581
526
474
585
486
394 I 6.8 to 1
5.1 to 1
4.1to1
power at fuel to
air ratio of 0.014

TABLE ID. - IDLE CONDITIONS

Model A- 1 Model A-2

Combustor efficiency, percent 92.4

Total-pressure loss, kPa 7.15
Pattern factor 1.12
Hydrocarbon emission index, g/kg 62.9
Carbon monoxide emission index, g/kg 96.9
Oxides of nitrogen emission index, g/kg 3.9
Smoke number 23.8

13
 TABLE IV. - EXPERIMENTAL PERFORMANCE AND EMISSIONS DATA

Airflow Combustor Combustor Combustor Reference Fuel-air Fuel pres- Pattern Gas sample Exhaust emissions index Smoke
rate, inlet tem- inlet pres- pressure velocity, ratio sure drop, factor combustion number
kg/sec perature, sure, drop, m/s kFa. efficiency, Oxides of Hydro- Carbon Ratio of
K kPa percent percent nitrogen carbons mon- fuel-air
oxide ratios

Model A- 1

1.39 5.49 0.0095 94.47 0.23 97.8 7.3 I .I 65.6 1.020 13.51
1
1.58 5.57 ,0119 156.64 .19 98.9 I .3 2.6 40.8 1.024 15.1 I
1.57 5.37 ,0119 154.80 .19 98.9 8.0 2.2 39.6 1.029 -___
2.251
2.213

2.239 ,
2.250
I 685.80
685.11

695.22
681.66
E 1022.5
1023.9
1039.6
1.63
1.65
5.52
5.34
,0140
.0161
235.94
303.92
.22
.20
99.6
99.8

91.9
99.7
12.0
13.0

6.5
11.5
.6
.3

7.7
.9
15.5
7.0

51.1
8.8
1.050
1.014

1.054
1.059
12.3
14.5

20.6
16.4
I

'
2.283 682.46 1019.6 , 98.0 5.2 I. 1 55.9 ,997 13.5 I
2.283 688.60 993.2 99.2 8.3 1.1 28.8 1.006 17.8
2.260 679.82 1007.5 99.6 10.8 1
1
.6 13.0 '
I
1.016 11.0 '

2.285 682.78 1002.9 1.40 5.59 0.0098 ~ 103.83 ' 0.36 I 97.3 ' 7.2 I 8.8 11.5 0.963 ----
2.285 689.80 1024.4 1.46 5.53 ,0099 105.62 .34 1 98.6 , 8.5 3.8 43.9 1.024 26.0
2.275
2.253
681.40
683.49
1021.9
1011.4
1.59
1.47
5.45
5.47
.0096 91.80 ~ .32 I 91.7 , 1.5 i 1.2 61.0 ,999 17.0
.0143 240.65 .28 99.1 ' 10.5 .3 10.9 1.029 20.2
3.002 102.95 1312.4 1.38 5.53 .0099 200.66 .21 I 99.4 9.9 1 1.3 19.1 1.024 19.5
I
2.994 , 703.04 1366.2 1.59 5.54 0.0121 315.41 0.22 99.8 12.7
~

0.4 ' 1.6 1.032 21.2 '

3.029 702.10 1334.0 1.73 5.13 ,0141 444.99 .24 99.9 14.1 .2 3.4 1.039 17.0
3.663 ' 112.39 1102.4 1.41 5.51 ,0098 308.34 .21 99.8 12. I .4 7.0 1.035 26.0
3.634 113.95 1615.0 ' 1. 68 5.77 ,0123 497.56 .21 99.9 15.0 .2 2.5 1.022 , 24.5
3.654 718.40 1617.6 1.69 5.83 ,0138 647.60 .21 100.0 16.2 .1 1.5 1.038 19.3

3.554 109.11 1616.4 1.21 5.61 0.0101 261.91 1.33 99.8 12.7 0.3 7.1 1.004 28.2
3.564 705.58 1641.8 1.06 I 5.51 ,0104 83.15 1.42 99.7 1 2 . 5 , l . l 1.1 1.043 30.0
3.561 101.41 1638.1 1. 16 5.53 ,0125 146.93 .37 99.9 15.2 .1 ! 2.7 1.064 31.2
3.606 115.54 1642.9 .93 5.65 .0101 315.06 ' .43 99.8 13.7 .3 5.9 1.008 28.2
3.599 714.95 1602.0 1.41 ' 5.77 .0123 505.95 .41 99.9 16.0 .1 2.4 1.027 27.2

3.670 711.81 1666.2 1.33 5.68 0.0138 683.14 0.31 100.0 _____ 0.1 1.3 1.042 ----
2.081 625.53 865.8 1.76 5.42 ,0099 81.86 .I3 78.3 4.1 ~ 111 ~ 199 ,972 11.5
2.094 621. 08 862.3 2.08 5.42 ,0144 203.01 .35 98.8 6.1 2.8 i 38 .988 24.0 I

1.812 577.23 714.0 1.59 5.26 ,0140 138.20 .54 j 19.9 2.6 151 189 ,956 22.4
I
1.813 518.72 106.5 1. 96 5.34 .0122 100.16 i 1.14 65.9 2.6 305 151 .I94 15.0

1.522 532.10 533.1 ' 2.33 5.46 , 0.0141 i _____ 4.2 25.1 1.199 _-__
I 1.533
1.247
524.67
485.08
533.3
413.4
2.13
1.93
5.42
5.26
! ,0140
i ,0196
93.61
91.31
115.90
!
:
2.03
1.80
1.14
~ 98.9
45.4 1.1
1.3
526
236.4
86
113.6
.483
,699
, 8.4
24.2
t 1.244 479.11 391.2 1. 96 5.48 I .0164 12.40 ~ 1.80
i 72.5
62.5 1.3 341. 1 120.5 .6ll , 20.0 1
. . 9 16.6 1.016 9.2
2.988 , 681.81 1011.1 2.90 1.30 0.0124 318.69 0.25 99.1 10.0 0.6 12.6 1.013 15.0
3.942 699.03 1383.0 2.58 1.16 ,0117 544.05 .20 99.9 12.2 .2 4.0 1.056 16.6
4.011 691.82 1355.1 2.80 1.42 ,0136 153.44 .22 99.9 13.2 .1 2.3 1.043 15.5
4.549 113.45 1640.9 2.38 1.11 .0121 183.04 .24 99.9 14.8 .2 2.0 1.051 20.0
4.569 112.04 1634.2 2.59 1.15 ,0138 1049.40 .22 100.0 14.9 .1 1.3 1.044 14.0

3.183 685.62 1011.6 4.65 9.22 0,0100 282.31 0.30 99.6 7.5 1.0 13.5 1.011 9.6
3.779 685.09 1030.7 4.11 9.03 ,0122 431.89 .93 99.8 9.5 .4 8.4 1.001 10.0
3.166 682.96 995.9 4.12 9.28 ,0139 718.91 .43 99.9 11.2 .l 4.4 1.019 9.5
3.110 689.48 1077.1 3.41 8.67 ,0119 515.12 .57 99.8 10.8 .2 6.5 1.004 10.5
3.112 685.35 1004.7 4.22 9.25 ,0120 513.65 .41 99.8 6.0 .3 1.5 1.009 i ----
3.770 681.47 987.5 3.16 9.43 0.0099 342.13 0.41 99.6 8.7 0.8 12.9 1.005 6.5
4.822
4.866
699.41
698.11
1340.0
1368.5
1380.2
1365.4 ,
,
3.49
2.98
3.10
3.49 ,
9.05
8.93
8.98
8.94
1 ,0102
,0101
,0120
,0141 I
482.91
622.60
913.42
1259.1
.28
.38
.44
.41
99.8
99.8
99.9
100.0 1
10.3
10.8
12.1
12.4
.4
.4
.3
.2 I
1.3 ,988
5.81.003
3.4 1.014
2.3 .999
13.6
20.5
19.5
19.3

1613.4 3.13 8.99 0.0103 123.10 0.241 99.9 12.2 0.2 3.1 1.014 19.0
1632.5 3.35 8.78 .0104 840.61 .34 99.9 12.2 .3 3.6 .966 25.0
1646.3 3.28 8.82 ,0124 1256.60 .40 99.9 12.8 .2 2.2 .986 21.3
1627.6 3.90 9.05 ,0142 1111.80 .44 100.0 13.6 .2 1.2 1.004 23.0

1049.9 1.26 5.32 0.01Zl 628.81 0.77 99. 9 12.5 0.2 3.1 0.940
1006.1 1.37 5.51 ,0102 440.90 .62 99.1 10.4 .I 8.5 .920 ---_
1341.5 1. 41 5.11 .0091 945.25 .45 99.9 14.1 .2 1.8 1.064 33.5
1360.6 1.38 5.60 ,0118 1391.20 .63 99.9 15.9 .1 1.3 1.043 34.1
1344.0 1.21 5.64 ,0101 110.16 .68 99.9 13.1 .1 2.4 ,690

1338.9 1.51 5.69 0.0121 1140.90 0.68 99.9 15.0 0. 1 1.8 0.860
1575.1 1.45 5.88 ,0103 1394.50 .50 99.9 6.8 .2 1.3 ,988
1615.9 1.38 5.13 .0116 1926.20 .51 99.91 7.2 .1 .9 1.071
1600.3 1.55 5.83 ,0100 1122.60 .82 99.9 14.9 .2 1.9 .880
401.5 1.16 5. 19 ,0103 124.12 1.12 92.4 3.9 62.9 96.9 .980

868.4 1.63 5.49 0.0135 863.48 0.44 99.9 10.0 0.2 3.81.008 50.1
865.4 1.51 5.44 ,0131 691.01 .66 99.9 10.1 .3 5.4 ,950 ----
669.1 2. 01 5.66 ,0131 645.36 .40 99.5 8.0 1.7 13.9 ,992 34.6
686.1 1.66 5.44 ,0143 531.35 .56 99.6 7.8 1.3 13.1 ,920 ----
511.1 1.97 5.67 ,0136 434.93 I .38 97.9 6.1 12.6 31.3 .982 32.8

1.514 521.39 521.2 1. 69 5.51 0.0123 244.21 2.8 309.0 43.1 0.830 ----
1
1
1.243 476.02 401.0 1.84 5.23 ,0141 211.66 4.3 45.2 ~ 68.5 .936 23.6
1.207 484.33 403.8 1.I O 1.06 76.3 2.1 289.0 58.1
1 .I80 ----

1 1 1
3.133 682.01 1009.6 3.I O .48 99.6 11.7 .1 3.1 ,996 19.0
3.782 680.35 1014.4 3. I2 .46 99.9 10.5 .2 4.4 1.045 21.0 1
4.901 691.05 1349.1 3.40 0.49 99.9 12.1 0:; 3.2 1.045
5.849 711.86 1594.4 3.30 .39 99.8 12.6 3.2 1.072
 --

Preheater Exhaust gases from four

exhaust J47 combustor cans

t 1 ,- Airflow control valve

r A i r orifice
Atmospheric or
altitude exhaust
Indirectly fired
heat exchanger a i r supply
’
I

”I I I Pressure
Automatic
a i r temperature
control valve
JI ’
,r Exhaust
control
valve

. I ~ -I, Water

Test 1 ; 1 L Water sprays 1’

iombustor 1 1
Fuel Temperature.
injectors J
omissions. and
smoke sample
probes
Figure 1. -Test facility.

Combustor center Iine

-- - _~

1 ,-Instrumentation
station c
I

station
Instrumentation
B
1 /I !I 21.5
(a) Cross section.
! o Liner thermocouple locations

Figure 2 -Test combustor. All dimensions in centimeters.

16
 Airflow

/'
CD-32241-07

(b) Isometric view.

0 Operable fuel injectors

0 Nonoperable fuel injectors

Model A - i Model A-2 Model A-2

(even ) (odd)
(c) Fuel injector density and location.
Figure 2 - Concluded.

17

II II I II 11.111111.11 I 111111111 11111 I I1 11111 1 .I11 11111.111 I 111 I

 Id) Aft view of combustor. (b) Combustor liners,

(c) Housing.
Figure 3. - Reverse-flow combustor.

18

-.. -. . ...... ... . .... ,, ,,. .,,...,,.-.. I. I

 0 Total pressure

combustor^
0 Static pressure
0 Temperature
Station 2
n Station 3 5 4 0 Gas sample probe

nn , ,, ,I
-1
,- -2

.__

12 Temperature rakes

TOP

,-Gas sample
9 Q1; probes (4)

180'
Station 2 Station 3 Station 4 Station 5
Viewing upstream Viewing upstream Viewing upjtream Combustor exit
Viewing upjtream
Figure 4. - Research instrumentation.

19
co
0

95r 0 Sea-level takeoff

0 Low-altitude cruise
A High-altitude cruise
5~10-~ 0 Model A-1
90- 0 Model A-2
Solid symbols denote isothermal
values
I 0
c
I -
a
a
4
a

5 85-1
gn o .010 .OH .012 .013 .014 .015 .016 .017
Fuel-air ratio
0
c
m (a) Effect of fuel-air ratio for combustor model A-1.

A-1 1 F
u iwuei A-2
I 1
80 I 01
I D .05 .06 .07 .08
Diffuser Mash number
.09 .10
I
Figure 6. - Reverse-flow combustor total pressure loss
over a range of inlet diffuser Mach numbers.
I

6o0 5 10 15 20
Simulated compressor pressure ratio
(b) Effect of reduced power conditions. Fuel-air ratio, 0.014.
Figure 5. - Combustion efficiency obtained with reverse-flow combustor operating with
simplex fuel injectors at simulated inlet flight conditions with Jet A fuel.
.7-

0 Sea-level takeoff
0 Low-altitude cruise
.6- A High-altitude cruise
Open symbols denote model A-1
Solid symbols denote model A-2

.5-

-
A A
.4-

.3t
0 U
n A
U
.2 a
A

.010 .Oll .012 .013 .014 .015 .016

(a) Reference velocity, 5.5 meters per second. Ib) Reference velocity, 9.1 meters per second.
Figure 7. - Outlet pattern factor over range of fuel-air ratios for simulated inlet flight conditions with Jet A fuel.
 0 Sea-level takeoff
Lw-altitude cruise
A High-altitude cruise
Open symbols denote model A-1
950r Solid svmbols denote model A-2

0 Model A-2
--- Required profile

8oo-x”-

7 50
Hub 1 1 ,009 .010 ,011 .012 .013 .014 .015 .016
.a .9 1.0 1.1 1.2 Fuel-air ratio
Ratio of average circumferential to average temperature
Figure 9. - Liner tem2eratureon inner turn of a reverse-flow combustor for
Figure 8. - Outlet radial profile for simulated sea-level takeoff inlet simulated inlet flight conditions over range of fuel-air ratios with Jet A
condition. Overall fuel-air ratio. 0.014with Jet A fuel. fuel.
 0 Model A-1
0 Model A-2

-
Reduced
power

I -1
12 16
Simulated compressor pressure ratio
(a) Effect of Dressure ratio at reference velocity of 5.5 meters per second.

0 Combustor model A-1

a Combustor model A-2

Sea-level Low-altitude High-altitude Sea-level Low-altitude High-altitude

takeoff cruise cruise takeoff cruise cruise
Reference velocity, 5.5 mls Reference velocity. 9.14 mls
(b) Effect of velocity,
-
Figure 10. Emission index of unburned hydrocarbons in reverse-flow combustor for constant fuel-
a i r ratio of 0.014with Jet A fuel.

23
 100 -
0 Model A-1
0 Model A-2

-
80

"\
r
(51
Y
m
I -
d
U
c
'-

.-c
0
60i \ Reduced
power
.-VI
E
W
E
E 4 0
E
c
0
-E
m
0

20

0
1D R

-
12

Simulated compressor pressure ratio

16 a

(a) Effect of pressure ratio at reference velocity of 5.5 meters per second.

0 Combustor model A-1

Combusto- model A-2

c
0
e
d
O
LL Sea-level
takeoff
Low-a ltitude
cruise
High-altitude
cruise
Sea-level
takeoff
Low-altitude
cruise
High-altitude
cruise

Reference velocity. 5.5 m/s Reference velocity, 9.14 m/s

(b) Effect of velocity.
-
Figure 11. Emission index of carbon monoxide in reverse-flow combustor for constant fuel-air
ratio of 0.014 with Jet A fuel.

24
 0 Model A-1
0 Model A-2

/
4
I I
8
I
12
I16
0
Simulated compressor pressure ratio
(a) Effect of pressure ratio at reference velocity of 5.5 meters per second.

0 Model A - l
Model A-2

" Sea-level Low-altitude Hiah-altitude Sea- -a Ititu de

takeoff cruise cruise takeoff cruise cruise
Reference velocity, 5.5 mls Reference velocity, 9.14 mls
(b) Effect of velocity.
Figure 12. -Emission of oxides of nitrogen in reverse-flow combustor for constant fuel-air ratio
of 0.014 with Jet A fuel.

25
 O'r 0 Model A-1
El Model A-2

L
Reduced
power
*- ...

O4 I.
8
.

Simulated compressor pressure ratio

Figure 13. - Smoke number obtained with reverse-flow combustor for .
constant fuel-air ratio of 0.014 with Jet A fuel.

26
 i
1400-
0 Model A-1
0 Model A-2

1200.

0 0
1GOO -

m
a
Y

ci 8M) -
e
u

600 -

400 -

200 -

Figure 14. - Simplex pressure-atomizing fuel injector pressure drop over range of fuel flows with
Jet A fuel.

27
 NASA Tp-1260
Report No.
AVRADCOM TR78-22(PL)
4. Title and Subtitle
I 2. Government Accession No.

~~
3. Recipient's Catalog No

5. Report Date
REVERSE-FLOW COMBUSTOR FOR SMALL GAS TURBINES July 1978
~-
6. Performing Organization Code
WITH PRESSURE-ATOMIZING FUEL INJECTORS
~~

7. Author(s)C a r l T. Norgren, Edward J. Mularz, and 8. Performing Organization Report No.

Stephen M. Riddlebaugh E-9458
10. Work Unit No.
9. Performing Organization Name and Address 5 05- 04
NASA Lewis R e s e a r c h Center and 11. Contract or Grant No
AVRADCOM R e s e a r c h and Technology Laboratories
Cleveland, Ohio 44135
13. Type of Report and Period Covered
12 Sponsoring Agency Name and Address
Technical Paper
National Aeronautics and Space Administration
Washington, D. C. 20546 and U. S. Army Aviation R e s e a r c h and 14. Sponsoring Agency Code
Development Command, St. Louis, MO 63166
I5 Supplementary Notes

6. Abstract
A reverse-flow combustor suitable for a s m a l l gas turbine (2 t o 3 kg/s m a s s flow) was used t o
evaluate the effect of pressure-atomizing fuel injectors on combustor performance. In t h e s e
tests a n experimental combustor was designed t o operate with 18 simplex pressure-atomizing
f u e l injectors at sea-level takeoff conditions. T o improve performance at low-power condi-
tions, fuel was redistributed so that only every other injector was operational. Combustor
performance, emissions, and liner temperature are compared over a range of p r e s s u r e and
inlet-air t e m p e r a t u r e s corresponding t o simulated idle, cruise, and takeoff conditions typical
of a 16 t o 1 p r e s s u r e r a t i o turbine engine.

7. Key Words (Suggested by Author(s1 1 18. Distribution Statement

Reverse- f low combustor Unclassified - unlimited
Fuel injection density STAR Category 07
Low power setting performance

9. Security Classif. (of this report)

Unclassified
20. Security Classif. ( o f this page]
Unclassified
1 21. ~ 0 Pages
~ ;

1 22. Price*
A03
-~ - -

NASA-Langley, 1978
. National Aeronautics and THIRD-CLASS BULK R A T E
Postage and Fees Paid I

(S)
' Space Administration National Aeronautics and
Space Administration
NASA451
Washington, D.C.
20546 USMAIL

Official Business
PenaltV for Private Use, $300

POSTMASTER: If Undeliverable (Section 1 5 8

Postal Manual) Do Not Return

-I- .- " -11 II " 1 1 I111-111 I I 1I. 1111 I 1111111 II "

1
. II 11111.1 IIII 1111I II 111-
. ..I
C' I1