PAXVE, INC.
JULY, 1971
FINAL REPORT
EVALUATION OF A LOW NOx BURNER
BY
E. B. ZWICK
T. R. MILLS
R. FIO RITO
PREPARED FOR
DIVISION OF ADVANCED AUTOMOTIVE POWER
SYSTEMS DEVELOPMENT,
OFFICE OF AIR PROGRAMS
ENVIRONMENTAL PROTECTION AGENCY
PAXVE REPORT USG-1
CONTRACT EHS 70 -125
MtfR
INC.
840 PRODUCTION PLACE • NEWPORT BEACH, CALIFORNIA 92660

-------
3BG/S inc.
840 PRODUCTION PLACE • NEWPORT BEACH, CALIFORNIA 92660
EVALUATION OF A LOW NOx BURNER
FINAL REPORT
For the period of June 29, 1970 thru July 28, 1971
By
E. B. Zwick
T. R. Mills
R. FioRito
PAXVE REPORT USG-1
The work upon which this publication is
based was performed pursuant to Contract
No. EHS 70-125 with the Division of
Advanced Automotive Power Systems
Development, Office of Air Programs,
Environmental Protection Agency.

-------
TABLE OF CONTENTS
DESCRIPTION
SUMMARY
OBJECTIVES OF THE PROGRAM
CONCLUSIONS AND RECOMMENDATIONS
EXPERIMENTAL METHODS
A.	Experimental Installations
B.	Instrumentation
C.	Test Procedures and Techniques
D.	Emission Data Collection &
Data Reduction
EMISSIONS MEASURING TECHNIQUES
A.	Description & Operation of
Instruments Used
B.	Instrument Calibration
C.	Emission Measuring Problems
EXPERIMENTAL RESULTS
A.	Experimental Data Listings
B.	Fuel/Air Ratio Analysis
and Correlation
C.	Experimental Emissions Data
D.	Experimental Stability Data
E.	Detailed Emissions Investigation
ANALYTICAL INVESTIGATION
A.	Literature Survey
B.	Burner Analysis
C.	Computer Analysis
CORRELATION OF DATA WITH THEORY
A.	Correlation of the Experimental
Stability Data
B.	Correlation of the Oxides of
Nitrogen Data
C.	Correlation of the CO Emissions
Data

-------
1
2
3
4
ON
1
2
3
4
5
6
7
8
9
10
11
12
13
14
IS
ON
1
2
3
4
5
ILLUSTRATIONS
PAXVE BURNER STABILITY DATA - PROPANE - AMBIENT
PAXVE BURNER OXIDES OF NITROGEN DATA
PAXVE BURNER CARBON MONOXIDE DATA
PAXVE BURNER HYDROCARBON EMISSION DATA
BURNER EVALUATION FACILITIES
TEST FACILITIES
BURNER TEST FACILITIES
TEST FACILITY FOR GAS EMISSIONS ANALYSIS
TEST STAND 1 BURNER SCHEMATIC
BURNER TEST STAND 2, SCHEMATIC DIAGRAM OF FUEL AND AIR
SYSTEMS
FUEL HEATER - VAPORIZER - BURNER FUEL SYSTEM
ELECTRICAL FUEL VAPORIZER
BURNER TEST STAND 2, SCHEMATIC DIAGRAM OF VAPOR GENERATOR
SYSTEM
BURNER VAPOR GENERATOR ASSEMBLY
FUEL-AIR COMBUSTION DATA
OXYGEN SYNERGISM EFFECT ON THE FLAME IONIZATION DETECTOR
EMISSIONS NORMALIZING FACTORS FOR PROPANE-AIR
EMISSIONS NORMALIZING FACTORS FOR OCTANE-AIR
BURNER STAND No. 1 - TABULATION OF SUBSYSTEMS S CONTROLS
BURNER STAND No. 2 - TABULATION OF SUBSYSTEMS & CONTROLS
BAILEY HEAT PROVER WITH OPERATING SCHEMATIC
VOLUMETRIC GAS ANALYSIS APPARATUS
CALIBRATION CURVE FOR DETERMINING OXIDES OF NITROGEN
CONCENTRATION BY GRIESS SALTZMAN METHOD, LOW RANGE
CALIBRATION CURVE FOR DETERMINING OXIDES OF NITROGEN
CONCENTRATION BY GRIESS SALTZMAN METHOD, HIGH RANGE
MODEL 8004 GAS CHROMATOGRAPH USING THERMAL CONDUCTIVITY
DETECTOR DETECTING CARBON DIOXIDE & CARBON MONOXIDE

-------
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
ILLUSTRATIONS (Cont.)
INTERNAL CONFIGURATIONS OF GAS CHROMATOGRAPH DETECTORS
PEAK HEIGHT OPTIMIZATION FOR CO ELUTION FROM GAS CHROMATOGRAPH
CALIBRATION CURVES FOR GASES USED WITH THE FLAME IONIZATION
DETECTOR
CARBON MONOXIDE CALIBRATION ON THERMAL CONDUCTIVITY DETECTOR
GAS CHROMATOGRAPH
CARBON DIOXIDE CALIBRATION ON THERMAL CONDUCTIVITY DETECTOR
GAS ChROMATOGRAPH
OXYGEN CALIBRATION ON THERMAL CONDUCTIVITY DETECTOR GAS
CHROMATOGRAPH
CALIBRATION OF SEPARATION SIDE OF FID GAS CHROMATOGRAPH
SCHEMATIC DIAGRAM OF EXPONENTIAL DILUTION APPARATUS
PEAK HEIGHT FROM CHROMATOGRAM, SCALE DIVISIONS - L-9000
SENSITIVITY TEST 3
ABOVE - TEST 2
CALIBRATION OF FLAME IONIZATION DETECTOR SPAN GAS DILUTED
WITH NITROGEN
CALIBRATION OF FLAME IONIZATION DETECTOR SHOWING OXYGEN
SYNERGISM EFFECT SPAN GAS DILUTED WITH ZERO AIR
COMPARISON OF INITIAL & FINAL DATA
NOx SAMPLING POSITIONS
CALIBRATION OF GRIESS-SALTZMAN ABSORBING REAGENT
KEROSENE DEW POINT DATA
DIAPHRAGM PUMP USED TO OBTAIN HYDROCARBON DATA - BURNER
EXHAUST SAMPLING SYSTEM
OXYGEN SYNERGISM EFFECT ON VARIOUS HYDROCARBONS USING PURE
HYDROGEN & AIR
OXYGEN SYNERGISM EFFECT ON VARIOUS HYDROCARBONS USING A
NITROGEN-HYDROGEN BLEND AND AIR
OXYGEN SYNERGISM EFFECT ON VARIOUS HYDROCARBONS USING A
HELIUM-HYDROGEN BLEND AND AIR

-------
ILLUSTRATIONS (Cont.)
SECTION V
TABLES
1	COMPARISON OF NOx SAMPLING POSITIONS
2	TEMPERATURE & AGING EFFECTS ON GRIESS-SALTZMAN ABSORBING AGENT
3	EFFECT OF EVACUATING PROCEDURE ON SAMPLING RESULTS
4	FUEL SPECIFICATION - KEROSENE
SECTION VI
TABLES
1-13	EXPERIMENTAL DATA FROM THE PAXVE BURNER
14	NOMENCLATURE FOR EXPERIMENTAL DATA TABLES
15	SIGNIFICANT TEST PROGRAM MILESTONES
16	- 23	COMPARISON OF FUEL AIR RATIO VALUES
24	FUEL AIR RATIO CORRECTION FACTORS FOR FLOWMETER DATA
25	- 32	THEORETICAL FLAME TEMPERATURES
Fig. 1	FUEL-AIR COMBUSTION DATA
" 2	VOLUMETRIC OXYGEN DATA
3	COMPARISON OF VOLUMETRIC FUEL/AIR RATIO VALUES
n 4	CARBON DIOXIDE VALUES VERSUS NOMINAL FUEL/AIR RATIO
" 5	VOLUMETRIC OXYGEN DATA VERSUS NOMINAL FUEL/AIR RATIO
6	CARBON DIOXIDE VALUES VERSUS CORRECTED FUEL/AIR RATIO
" 7	CARBON DIOXIDE VALUES VERSUS FLOW METER FUEL/AIR RATIO
8	CARBON DIOXIDE DATA FROM THE GAS CHROMATOGRAPH
9	COMPARISON OF CARBON DIOXIDE CIIROMATOGRAPH & VOLUMETRIC DATA
" 10-18	PAXVE BURNER EMISSIONS
" 19	CORRELATIONS OF OXIDES OF NITROGEN DATA
" 20-26	PAXVE STABILITY DATA
n 27-37	CO EMISSIONS DATA FROM THE PAXVE BURNER
" 38-48	COG EMISSIONS DATA FROM THE PAXVE BURNER
" 49-55	HC EMISSIONS DATA FROM THE PAXVE BURNER
n 56-64	HCG EMISSIONS DATA FROM THE PAXVE BURNER
" 65-82	NOX EMISSIONS DATA FROM THE PAXVE BURNER

-------
1
3
4
5
6
7
8
9
10
11
12
13
14
20
21
IS
ON
1
3
4
5
6
7
ILLUSTRATIONS (Cont.)
SEMENOV'S THERMAL IGNITION THEORY
SIMPLIFIED COMBUSTION MODEL
VULIS' COMBUSTION THEORY
EFFECT OF INLET TEMPERATURE ON CRITICAL PHENOMENON
EFFECT OF FLOW RATE ON CRITICAL PHENOMENON
COMBUSTION SYSTEM WITHOUT CRITICAL (IGNITION & EXTINCTION)
PHENOMENON
FLAilE STABILITY CURVE, GUTTER BURNER
"	"	" CAN BURNER
CAN BURNER & GUTTER BURNER CONCEPTS
RECIRCULATION MODEL
BURNER ANALYSIS TECHNOLOGY
STABLE OPERATING HEAT BALANCE
INCIPIENT BLOWOUT HEAT BALANCE
THEORETICAL BURNER ANALYSIS
BURNER STABILITY LIMIT ANALYSIS
COMBUSTION INTENSITY PARAMETER, LOG PLOT
THEORETICAL BURNER ANALYSIS
BURNER STABILITY CORRELATION
CORRELATION OF OXIDES OF NITROGEN DATA WITH RECIPROCAL
COMBUSTION TEMPERATURE
CORRELATION OF OXIDES OF NITROGEN DATA - KEROSENE - HOT -
BURNER DATA
CORRELATION OF BURNER OXIDES OF NITROGEN DATA
COG EMISSIONS DATA FROM THE PAXVE BURNER
PAXVE BURNER CARBON MONOXIDE DATA

-------
ILLUSTRATIONS (Cont.)
SECTION VIII
TADLES
1	PROGRAM PREDICT
2	PROGRAM LIMIT
3-18	COMPARISON OF PREDICTED & EXPERIMENTAL BURNER STABILITY DATA
19-31	COMPARISON OF PREDICTED BURNER INEFFICIENCY & EXPERIMENTAL
DATA

-------
I. INTRODUCTION Ai7D SUMMARY
A.	Introduction
During the past twelve months, Paxve, Inc. of Newport
Beach, has been engaged in an experimental and analytical
investigation of the Paxve burner, a low emission combustion system.
The Paxve Burner is a proprietary device developed by Paxve in
conjunction with work on an automotive Rankine cycle engine. The
burner in its present form consists of a structure embodying certain
fuel injection and combustion concepts, together with a set of
operating conditions which allow low emission operation.
Preliminary emission surveys conducted by Paxve during
the summer and fall of 1969, showed the unusually low emission
characteristics of the Paxve Burner, particularly with reference
to oxides of nitrogen. Correlations of the NOx data at that time
showed that the NOx emission levels could be correlated with the
operating temperature of the burner. Those preliminary results also
suggested that the low NOx was a benefit derived from the wide
stability limits of the burner.
On June 29, 1970, Paxve entered into a contract with the
Division of Advanced Automotive Power Systems Development, Office of
Air Programs, Environmental Protection Agency. The purpose of that
contract was to investigate the emission and stability character-
istics of the Paxve burner. Experimental work on the program was
completed early in May, 19 71. That experimental work together with
analytical investigations of the burner and correlation of the
experimental data form the subject of this final report.
B.	Summary
The results of the experimental and analytical invest-
igation conducted herein may be summarized as follows:
1. Program
Measurement of emissions from the Paxve burner
including oxides of nitrogen, carbon monoxide and unburned hydro-
carbons were made as a function of burner operating conditions.
Variables investigated during the program included burner air flow
rate, fuel/air ratio, air and fuel inlet temperature, and burner
volume. Stability characteristics of the burner were investigated
using both propane and vaporized kerosene as fuel. Emission charac-
teristics of the burner were determined for both of these fuels over
a range of mixture ratios ranging from lean blowout to rich blowout.
(Rich operation at high flow rates was limited by the fuel handling
capability of the facility.) Emission measurements were made
directly from the burner exhaust. Additional measurements were also
made downstream of a vapor generator operating in conjunction with
the burner. The vapor generation loop was one in which an organic
working fluid was circulated through a helical coil over which the
burner exhaust passed. An approximation of the influence of vapor

-------
generator quenching on burner emissions was thus determined.
A theoretical analysis was conducted of burner operation
which provided predictions of burner stability limits and the
efficiency of the burner operation under stable operating conditions.
That analysis was used as an aid in correlating the experimental
data.
2.	Burner Stability
Stability of the burner was found to be closely
within the limits predicted by the theoretical analysis. Figure 1
shows a typical set of burner stability data with the corresponding
theoretical lean blowout limit. The influence of air flow rate,
air temperature, and burner volume showed good agreement with the
theoretical blowout limits predicted by the theory for lean
operation. Some rich blowout data was obtained, the theoretical
analysis did not treat this case and therefore, no comparsion with
theory here is possible.
The theoretical analysis predicted somewhat wider blowout
limits at very low flow rates than were experimentally observed.
This has been attributed to the failure of the theory to account
for heat loss from the burner which can be significant at low flow
rates. The Paxve burner is a very low heat loss device and hence
the deviations from theory m this regard were not large.
3.	NOx Emissions From the Burner
Figure 2 shows the influence of fuel/air ratio
and air flow on the oxides of nitrogen emissions from the Paxve
burner, burning vaporized kerosene with an air inlet temperature of
400°F. This data is typical of the experimental NOx data taken from
the burner after run 282 when a fuel injector problem was solved.
The oxides of nitrogen emissions fall below the 1975 EPA goal* of
1.38 gr/Kg for fuel/air ratios less than 80% of stoichiometric
(f/a = 0.051). The influence of the air flow rate on the NOx
emissions is minor and tends to be obscured by scatter in the data.
The influence of air inlet temperature is rather strong. When the
data is plotted against combustion temperature, however, the effect
of the inlet air temperature is greatly reduced.
Oxides of nitrogen levels increase with increasing fuel/
air ratio, approaching the theoretical equilibrium values at about
an equivalence ratio of 1.2 (f/a = 0.08). At low fuel/air ratios,
the levels are orders of magnitude below the equilibrium values.
Normal operation of the Paxve burner lies in a combustion
temDerature range of 2400° to 2700°F. The corresponding fuel/air
range of operation the burner generally has emission levels of less
than 0.2 gm/Kg. This is seven times better than the EPA goals.
* See paragraph 10 below

-------
4. CO Emissions From the Burner
Figure 3 shows the influence of fuel/air ratio
and air flow on the carbon monoxide emissions from the Paxve burner,
burning kerosene. The data in this curve are typical of the
experimental CO values reported here. At low fuel/air ratios carbon
monoxide emissions are determined primarily by burner efficiency
considerations. These in turn are quite sensitive to air flow per
unit volume, fuel air ratio, and air inlet temperature. Carbon
monoxide emissions at high fuel air ratios parallel the curve of
theoretical equilibrium concentration of CO in the burner exhaust.
CO emission levels can be kept below the 19 75 goal* of
16.2 gr/Kg of fuel by providing the proper combination of fuel/air
ratio, volume, air flow and air inlet temperature. In the normal
operating range of the burner (between f/a = 0.033 and f/a = 0.038)
carbon monoxide emissions are within acceptable range below air flow
rates of 180#/hr.
5.	Hydrocarbon Emissions From the Paxve Burner
Figure 4 shows hydrocarbon emissions from the
burner operating with vaporized kerosene and heated inlet air. It
is typical of the HC data from the burner. In the very low fuel/air
range near the burner stability limit hydrocarbon emissions from the
burner with kerosene as the fuel show a very strong dependence on
fuel/air ratio. In this region the HC emissions with kerosene are
somewhat influenced by burner efficiency considerations. A similar
increase of HC emissions near the lean stability limits is observed
with propane, but the hydrocarbon levels are much lower and the
appearance of any hydrocarbon is generally an indication of
incipient flame out.
At high fuel/air ratios hydrocarbon emissions climb to
relatively high values particularly beyond stoichiometric. Between
these two extremes there is a broad range of operating conditions for
which hydrocarbon emissions from the Paxve burner are essentially
zero. In this range of operation the emission levels were so low
that they could not be accurately measured with a flame ionization
detector.
Emission levels for hydrocarbons can be kept below the
1975 EPA goal* of 0.48 gr/Kg by the proper selection of fuel/air
ratio operating range. Within the desired operating range, of
f/a = 0.032 - 0.038, the burner HC emission levels were less than
1975 goal over the entire range of air flow per unit volume tested.
6.	Influence of the Vapor Generator on Emission
Level
Figures 2, 3, and 4 also show the comparsions
between the emissions from the vapor generator exhaust and the
burner. There was some reduction in the oxides of nitrogen
emissions between samples drawn from the burner (bottom of the
* See paragraph 10 below
1-3

-------
stack) and samples drawn from the vapor generator exhaust (top of
the stack). CO emission levels and hydrocarbon emission levels were
strongly influenced by the vapor generator. The low values of
hydrocarbon and CO emissions which occur in the mid-range of fuel/
air ratios continued to occur with the vapor generator exhaust. The
increase in carbon monoxide and hydrocarbons which characterize the
burner exhaust at very low fuel/air ratios disappeared when samples
were taken downstream of the vapor generator exhaust. In examining
the CO curves, it should be noted that the limit of resolution of
the gas chromatograph was 5 ppm. Almost all of the vapor generator
CO data actually failed to show any measurable CO. These data
points were then recorded as 5 ppm. It appears that the net
influence of the vapor generator is to quench the NOx formation
reaction while permitting continued oxidation of the hydrocarbons
and carbon monoxide into carbon dioxide and water vapor.
The emissions measured in the vapor generator
exhaust fall below the EPA goals over a much wider range of
operating conditions than the emissions from the burner. The upper
limit on the allowable fuel/air ratio range in both cases is set by
the oxides of nitrogen levels. Fuel/air ratios below f/a = 0.05
are necessary to keep the NOx emissions below 1.38 gm/Kg. The low
fuel/air ratio limit for the burner is set primarily by the CO and
HC emissions since the NOx emissions continue to fall as the fuel/air
ratio is dropped. This lower limit is reduced considerably by the
presence of the vapor generator. If this result persists in combin-
ation with a vapor generator designed for an automobile, the Paxve
burner will be capable of providing very low emission levels over an
extremely wide range of operating conditions.
7.	Effect of Non-Uniform Fuel Distribution
The Paxve burner described here used propane or
vaporized kerosene as its fuel. The fuel and the air were premixed
in the inlet pipe before entering the burner. Most of the informa-
tion described in this summary is based on the behavior of the
burner with well mixed homogeneous inlet flow which was achieved
after run 282. Earlier experiments with the burner included some in
which there was an unsuspected maldistribution of the fuel/air
mixture in the inlet pipe. With severe maldistribution of the fuel,
burner emission levels of all types were found to increase signif-
icantly. Most startling in this regard were the very high levels of
hydrocarbons (on the order of 30 ppm) measured downstream of the
vapor generator exhaust. The oxides of nitrogen and carbon monoxide
emissions from both the burner and vapor generator exhaust were
also higher when the improper fuel air distribution existed. All
of these problems disappeared as soon as the fuel injection
problem was discovered and corrected.
8.	Theoretical Analysis
A simplified well stirred reactor model of
burner operation was used to provide estimates of burner stability
limits and efficiency as a function of the operating conditions.
Variables in the analysis included the volume of the burner, the
air flow rate, the equivalence ratio, and the air inlet temperature.
1-4

-------
The anlaysis followed well established lines of theoretical burner
analysis.
It was found that the burner stability data correlated
very well with the blowout limits predicted by the burner theory.
It was further found that the carbon monoxide and hydrocarbon
emission levels were considerably below those predicted by the
theoretical analysis, but they followed the same trends and were
influenced by the expected variables.
Oxides of nitrogen emission show a strong correlation
with temperature, and are less influenced by air flow levels than
had been expected. Efforts to provide a theoretical model for
correlating the oxides of nitrogen data have been only partially
successful.
9. Problems of Emissions Measurement
Problem areas were encountered in all of the
emission measurements, including oxides of nitrogen, carbon
monoxide, and hydrocarbons. All of these problems were resolved and
satisfactory data obtained which is reported herein. Some of the
problems encountered deserve special recognition and more investi-
gation .
Oxides of nitrogen data include both NO2 and NO. The NO2
is highly soluable in water. The NO is unstable and readily oxidizes
to NO2• At very low NOx emission levels, it is important not to
lose the NO2. This in turn requires special care in handling the
water vapor which is produced by the combustion process. Many of
the presently accepted standard techniques involve separation of
the water in a fashion which can trap the NO2 before it is measured.
This may mean a loss of as much as 50% of the total NOx in a lean
operating, low emission mode.
Carbon monoxide data from the Paxve burner was frequently
below 5 ppm, this was the lower limit of resolution of the gas
chromatograph which was used for this purpose. There are instru-
ments available which will read to lower levels, but these are not
common in the automotive research field.
The measurement of hydrocarbons with a flame ionization
detector proved generally satisfactory after a heated sample line
and heated sampling pump were installed. Two problems remained.
The first of these is the so called oxygen synergism effect. The
change m sensitivity of the instrument with varying oxygen content
in the exhaust stream was not only a nuisance in data reduction, but
is a potential source of error. The magnitude of the effect is a
function of the compositon of the stream, and hence the appropriate
correction can only be made if an accurate hydrocarbon analysis is
available. Most of those using these instruments are unable to
make such an anlaysis, or to use it properly since the magnitude of
the effect is not readily available for all of the combinations of
hydrocarbons which one might expect to encounter.
The most severe problem found in hydrocarbon emission
1-5

-------
measurement was closely related to the unusually low levels of
hydrocarbons which are found in the Paxve burner exhaust. We
frequently observed negative output from the recorder which was
monitoring the signal from the FID. The negative values were on
the order of -0.5 ppm expressed as hexane. These were not
spurious readings caused by drift in the instrument or recorder
zero setting. An extensive investigation of this problem showed
that it might be caused by the presence of water vapor in the
burner exhaust gas stream. Addition of 10% water vapor to a stream
of "zero" air caused a similar negative zero shift. This of course
represents the lower limit of resolution of the hydrocarbon
measurement capability of the FID used in this program. This zero
shift deserves further investigation. We have not seen it reported
elsewhere.
Combustion of hydrocarbon fuels with air can be achieved
with emission levels substantially below the goals set by the
EPA for 1975 automotive standards. Experiments with the Paxve
burner indicate that this burner is a device which is capable of
achieving these low emission levels.
10. 1975 EPA Emission Goals
The Division of Advanced Automotive Power
Systems Development, EPA, has established certain vehicle emission
goals to be met for hydrocarbons, carbon monoxide, and oxides of
nitrogen. These goals expressed in grams per mile are: 0.14 g/mi
of HC, 4.7 g/mi of CO and 0.4 g/mi of NOx expressed as NO2. Using
an assumed average fuel economy of 10 mi/gal over the Federal
Driving Cycle, we can derive emission goals based on grams of
pollutant per kilogram of fuel. The resulting values are: 0.48
g/Kg of HC, 16.2 g/Kg of CO and 1.38 g/Kg of NOx. While these
values of grams of pollutant per kilogram of fuel are only approxi-
mations to what is needed for an automobile, they provide a conven-
ient means of comparing burner emission performance with the
1975 vehicle standards. The phrases "EPA Goals" and "1975
standards" which are used extensively in this report refers to these
derived g/Kg values.
1-6

-------
PAXVE BURNER STABILITY DATA
Propane Ambient
0.03	Q.04
FUEL AIR RATIO - F/a
0.3
0.««	0.5	0.6
EQUIVALENCE RATIO - 0
0.7
1.0
0.1
0.01
Fuel Air Ratio-f/a
PAXVE BURNER
OXIDES OF HITROCKII yA*
AIR FLOW PROP. KER,
f/Hr.
Under 40	-j-
40-70	O
70 -120	A
120 -150	~
Over 150	o
FLAG AIR FUEL
TEMP. TEMP.
Cold	Cold
Hot	Cold
Cold	Hot
Hot	Hot
KEROSENE DATA
VOLUME 52.3 cuin,
AIR TEMP. OVER
250 °F
Figure
1-1
Figure
1-2

-------
PAXVE BURNER
CARBON MONOXIDE DATA
AIR FLOW
#/Hr.
Under 40	-j-
40-70	O
70 -120	A
120 -150	~
Over 150	O
0.1
0.02
0.03	0.0H
0.05
0.06	0.07
KEROSENE DATA
VOLUME 52.3 cu in
AIR TEMP OVER 250 °F
1975 STANDARD FOR CO 16.2 g/kg
NOMINAL FUEL AIR RATIO FAN
Figure
1-3
nA"VF BIT*"!"'*
r'-T"
AIR FLOW
PROP.
KER.
FLAG
AIR
FUEL
#/Hr.



TEMP.
TEMP.
Under 40
+
+
A
Cold
Cold
40 - 70
0
•
A
Hot
Cold
70 -120
A
A
A
Cold
Hot
120 -150
~
¦

Hot
Hot
Over 150
O
•



0.02
0.0M
o. or
FUEL AIR RATIO - f/a
Figure
I-H

-------
II. OBJECTIVES OF THE PROGRAM
The objectives of the research program described in this
report were to obtain data on the characteristics of the Paxve
burner, a low emissions burner. The characteristics to be evaluated
were the exhaust emissions and the stability limits of the burner.
These data were to be combined with a theoretical analysis of burner
stability and operating characteristics in order to provide a
thorough evaluation of this burner concept.
A. Experimental Investigation
The experimental investigation was designed to determine
(1) the lean and rich blowout limits (stability limits) of the
burner, and (2) the emissions characteristics of the burner over
a wide range of operating conditions.
1.	Stability Limits
The lean and rich blowout limits for the burner
were determined for a wide variety of operating conditions. The
variables tested included fuel and air temperatures, fuel and air
flow rates, and fuel types.
a.	Fuel Type
The stability limits were determined
for propane and kerosene.
b.	Air Temperature
The air temperature was varied from
ambient to 400°F.
c.	Fuel Temperature
The fuel temperature was varied from
ambient to 800°F.
d.	Air Flow Rate
The air flow rate was varied from 15 to
150 lbs/hr.
Sufficient data were taken near each blowout limit so that
gross extrapolations were avoided.
2.	Emissions Characteristics
Burner emission measurements were made over
the range of operating conditions used in the stability limits
study. Measurements on the burner exhaust were made over a fuel/
air ratio range from the lean blowout limit to the rich blowout
limit. Exhaust gas measurements included (1) total unburned

-------
hydrocarbons, (2) oxides of nitrogen, (3) carbon monoxide, (4)
carbon dioxide, and (5) oxygen.
A closed circulating working fluid loop was constructed
to permit burner operation in combination with a vapor generator.
Tha effect of the vapor generator quenching on emissions was
determined by sampling the burner gases as they were exhausted
from the vapor generator stack.
B. Analytical Effort
1.	Literature Survey
A literature survey was made of burner perfor-
mance and stability analyses. Background data was gathered to
assist in making an analysis of stability and emission character-
istics of burners.
2.	Analytical Study
The literature survey provided the basis for an
analysis of the stability and emissions characteristics of the
burner under study. The analysis investigated the effects of
various parameters on the stability limits and performance charac-
teristics of the burner. The burner was modeled using a simplified
stirred reactor concept.
3.	Data Analysis
The experimental data on stability limits and
emissions were correlated with the results of the analytical study.
Additional semi-empirical correlation of data not treated by the
burner analysis, was conducted.
II-2

-------
III. CONCLUSIONS AND RECOMMENDATIONS
A. Conclusions
The results of the experimental and analytical
program discussed in this report show that for the size burner
tested (approximately 100,000 BTU/HR), the Paxve burner is capable
of low emission operation. If these favorable emission char-
acteristics can be retained as the size is increased to a value
of heat release rate and turn down ratio that would be required
by a practical automotive propulsion system (approximately
2,500,000 BTU/HR, and at least 20:1), then the Paxve
burner should be capable of substantially bettering the qoals for
pollutants expressed in grams/kilogram established for burners
by the Division of Advanced Automotive Propulsion System
Development, EPA. The results of the analytical investigation
and the correlation of that analysis with the experimental
data permit a fairly accurate prediction of burner stability
and emission levels. Some correlation has been achieved for
emission levels of oxides of nitrogen which are not treated
directly by the burner theory.
In addition to these conclusions with regard to the
characteristics of the Paxve burner, we must also make some
observations about the problem of emission measurements. It
was clear that emission measurement is a difficult task and
that much remains to be done in order to establish emission
measurement procedures which give reliable and consistent data.
During the course of this program Paxve uncovered problem areas
in emission measurement which we have not seen treated in the
literature.
Conclusions from the program are presented below. The
subdivisions are as follows:
1.	Characteristics of the Paxve Burner
2.	Influence of the Vapor Generator on Burner Emissions
3.	Influence of Non-uniform Flow Distribution on
Emissions
4.	Theoretical Analysis
5.	Data Correlation
6.	Emission Measurement Problems
7.	Summary
1. Characteristics of the Paxve Burner
The characteristics of the Paxve burner as
investigated in the experimental program described here fall into
two general categories; the stability characteristics of the
burner, and the emission characteristics of the burner.
a. Burner Stability Characteristics
The experimentally determined
stability characteristics of the Paxve burner are presented in
Section VI of this report. The burner exhibits more or less

-------
conventional lean blowout and rich blowout behavior. Lean blowout
occurs at a fuel air ratio which depends on air flow rate and
air inlet temperature. At an air flow rate of 100 lbs/hr and
ambient inlet temperatures lean blowout occurs at approximately
50% of stoichiometric mixture ratio (f/a = 0.0 32 for propane). With
increasing inlet temperature the blowout inlet decreases. At
400°F inlet temperature the stability limit for propane at
100 lbs/hr of air flow is approximately 43% of stoichiometric
(f/a = 0.028). In both cases, lean blowout occurs at a combustion
temperaure of approximately 2150°F.
Rich blowout was more difficult to determine because of
the limited fuel flow capabilities of the Paxve test facility. At
50 lb/hr of air flow rich blowout occurs at approximately 0.17
fuel/air ratio which corresponds to about 2.7 times stoichiometric.
Stability data with kerosene shows blowout occuring at slightly
higher fuel/air ratios then with propane, but at about the same
equivalence ratio.
The emission behavior of the burner near the stability
limit is different from the two fuels. With propane, lean blowout
is characterized by relativly high carbon monoxide emissions, but
almost no unburned hydrocarbons are detectable. When the blowout
limit is finally reached, the hydrocarbon emissions start to appear
and begin to climb. Eventually the burner goes out without a change
in the operating conditions. Burning flame out under these
conditions can take as long as ten to twelve minutes to
occur.
With kerosene, hydrocarbon emission begins to appear before
the lean limit is reached. There are essentiallly no hydrocarbon
emissions from the burner down to about f/a = 0.038 with hot air and
kerosene. As the fuel/air ratios are reduced further, hydrocarbon
emissions start to appear, but reach steady values which remain
constant while the burner continues to operate stably. A further
reduction in fuel/air ratio causes another climb in the emissions
which again stabilize at some value. Finally as the blowout
condition is reached hydrocarbon emissions climb without leveling
off and gradually the burner temperature falls, the oxygen content
in the exhaust increases, and the burner goes out.
The presence of a significant amount of hydrocarbon
emissions at lean operating points close to the lean blowout limit
is a major difference between the behavior of the burner with
kerosene near the lean limit and the behavior of the burner with
propane 'under similar conditions.
Near rich blowout the burner emits considerable hydro-
carbons both with propane and with kerosene, although the kerosene
emissions seem to be generally higher at the rich operating
conditions.
b. Oxides of Nitrogen Emissions
Characteristic¥
Oxides of nitrogen emissions from the
III-2

-------
Paxve burner show a very distinct correlation with fuel/air ratio.
This can of course, be interpreted as being a strong correlation
with burner combustion temperature. The influence of inlet
temperature is very nearly what one might expect if a true
correlation with burner temperature were to apply. In the data
which is plotted in the ppm mode it is difficult to see any
pattern showing an influence of flow rate on the burner emissions
because of the data scatter. In general it appears that lower
flow ratios give somewhat higher emissions at a given fuel/air ratio.
For both kerosene and propane with ambient fuel and air
we can keep the emissions of oxides of nitrogen below 10 ppm
by maintaining the fuel/air ratio of the burner below 0.045
(approximately 70% of stoichiometric). At 400°F inlet temperature,
the 10 ppm value is reached at about f/a = 0.04 (0.625 equivalence
ratio).
The oxides of nitrogen data from the burner in lean
combustion fall far below the equilibrium NO concentrations which
would exist in equilibrium at the burner operating conditions.
At rich operating conditions of approximately 1.3 equivalence
ratio and higher (fuel/air ratio of = 0.08) the burner emissions
are approximately equal to the equilibrium values. In terms
of the emissions goals of 0.4 grams of NC>2 per Kg of fuel, we find
that the burner shows emission characteristics below this level
up to mixture ratios of approximately 70% of stoichiometric.
Maximum oxides of nitrogen measurements from the Paxve
burner were on the order of 110 ppm or 2.6 qrams per Kg of fuel.
Under these conditions, the burner was operating at or near
stoichiometric mixture ratio, an operating point which the burner
is not designed for and at which only very limited burner life could
be achieved. At its nominal operating condition of 2500°F flame
temperature the Paxve burner shows emissions levels on the order
of G.'l gm/Kg of oxides of nitrogen. This is a significant
improvement over the stated emissions goals of EPA.
c. Carbon Monoxides Emissions Data
Emissions of carbon monoxide from the
Paxve burner show two distinct trends. For mixture ratios above
f/a = 0.05 the carbon monoxide emissions are characterized by a
line of rapidly increasing emissions with increasing mixture
ratio which follows the equilibrium carbon monoxide concentration
in the burner exhaust. At mixture ratios below f/a = 0.0 35 the
carbon monoxide concentration in the exhaust follows a character-
istic of increasing CO emissions with decreasing mixture ratio. The
nature of these curves suggest a falling burner efficiency with
decreasing mixture ratio. Both families of curves show some
dependence on flow rate and a strong correlation with fuel/air ratio.
Increasing flow velocity increases the CO emissions in both cases.
We can maintain carbon monoxide emissions in the burner
exhaust below 16 gr/Kg of fuel, by providing an appropriate
combination of fuel/air ratio, air flow rate, and air inlet
temperature.
III-3

-------
At ambient inlet temperature, a flow rate of 50 lb/hr
will give less than 16 gr/Kg of CO at a fuel/air ratio above
0.036. Increasing the inlet temperature to 400°F decreases the
fuel/air requirement to 0.030 and 0.032 at 50 lbs/hr and 100
lb/hr respectively.
d. Hydrocarbon Emissions From the
Paxve Burner
Hydrocarbon emissions from the Paxve
burner exhibit some of the same characteristics as the carbon
monoxide emissions. At mixture ratios above about f/a = 0.050
hydrocarbon emissions increase with increasing mixture ratio,
increasing rapidly above stoichiometric. There is a wide range
of operating conditions for the burner between approximately
0.032 mixture ratio up to approximately 0.050 mixture ratio within
which the hydrocarbon emissions from the burner are zero to within
the limits of our measurement capability. At mixture ratios below
approximately 50% stoichiometric, hydrocarbon emissions again appear
at ambient inlet temperatures. With elevated temperatures, the
mixture ratio at which hydrocarbon emissions begin to appear from
the burner is somewhat lower.
The marked increase in hydrocarbon emissions at very
lean mixtures approaching lean blowout is characteristic of the
kerosene combustion data. With propane only minor hydrocarbon
emissions were observed as lean limit was approached until a limit
point had actually been reached. For propane combustion the
first significant appearance of hydrocarbons in the exhaust was a
sensitive indication of incipient lean blowout.
Hydrocarbon emissions can be kept within the nominal
limits assigned by EPA (0.48 gm/Kg) by an appropriate choice of
operating fuel/air range, at a given flow rate and inlet
temperature. At ambient inlet temperature and 50 lb/hr,
hydrocarbon emissions from the burner will be below 0.048 gm/Kg
above f/a = 0.035 with kerosene as the fuel. At 100 lb/hr a
slightly higher f/a value is required. When the inlet temperature
is increased to 400°F, the required minimum f/a for satisfactory
HC levels is about f/a = 0.032 at 50 lb/hr and 0.035 at 100 lb/hr.
The fuel/air limits necessary to meet the EPA hydrocarbon limita-
tions are wider than the f/a limits for satisfactory CO levels.
2. Influence of the Vapor Generator on Burner
Emissions
The vapor generator appears to have a major
effect on the CO and HC emission levels in the exhaust stream.
Oxides of nitrogen measurements downstream of the vapor generator
were in substantial agreement with the oxides of nitrogen
measurements upstream of the vapor generator. The influence of
the vapor generator on the CO and HC emissions is inferred primarily
from the plots of emissions versus fuel/air ratio. Simultaneous
measurements of CO were taken on only a few runs. No simultaneous
measurements of hydrocarbon emissions from the vapor generator
and burner were taken.
III-4

-------
a. Effect on NOx Emissions
NOx was measured in both locations
for a wide variety of runs.
On some runs there appeared to be a slight increase in
the NOx at the top of the stack compared to the burner, while on
other runs there appeared to be a slight decrease. The decrease
was frequently associated with conditions where the vapor
generator had not yet warmed up and some of the water vapor in
the burner exhaust was condensing on the vapor generator coils.
Previous experience with NOx measurements showed that NO2 dissolves
in condensed water from the exhaust, leading to a reduction in
the NOx level in the exhaust stream. The increased residence
time of the gases as they flow through the vapor generator does
not appear to significantly alter the NOx exhaust concentration.
b.	Effect on CO Emissions
Carbon monoxide emissions in the
exhaust from the vapor generator are generally much lower than the
emissions measured from the burner. We did not in general make
simultaneous measurements of these parameters hence it is
difficult to establish one to one correspondence between the levels.
An examination of the CO emissions as a function of mixture ratio,
however, and specific point by point comparisons indicate the extent
of this reduction. At the lean operating points where the CO
emissions tend to become excessive, the measurements in the vapor
generator exhaust were generally below the limit of the measuring
instrument.
Part of the difficulty in assessing the influence of
the vapor generator on the carbon monoxide emissions lies in the
very low levels of CO which are found in the Paxve burner exhaust
under normal operating conditions. These normal operating
conditions correspond to mixture ratios which give combustion
temperatures in the range of 2400°F to 2700°F. Testing outside
this range was limited to the burner alone. The structural
capabilities of the burner made it possible to do this. Testing
with the vapor generator loop in place, however, required
limitation of the combustion temperature range to avoid damage to
the working fluid passing the vapor generator loop. Within
this somewhat narrower temperature range, the CO emissions were
usually quite low, frequently below the ability of the
chromatograph to read, which was about 5 ppm. We have generally
identified the low levels of CO in our data which were otherwise
unreadable as being 5 ppm. Many of the CO values undoubtedly were
below this level. It would require sensitive instrumentation to
accurately establish the influence of the vapor generator on
the OC emission levels. It is reasonable to suppose, however,
that the reduction in temperature of the gases as they pass
through the burner is gradual enough to permit some recombination
of the CO with the available oxygen.
c.	Effect on HC Emissions
III-5

-------
The influence of the vapor generator
on hydrocarbon emissions was more difficult to determine than its
influence on carbon monoxide. The restricted operating range of the
burner when the vapor generator was in place resulted almost
entirely in zero or negative hydrocarbon readings, from the top
of the stack. Negative readings were obtained at conditions which
gave positive readings from the burner alone.
The negative readings, as explained elsewhere, are
probably due to the presence of water vapor in the exhaust which
causes a zero shift in the flame ionization detector. With zero
or negative readings, one has no reliable values that he can
use as a basis for an exact comparsion. We can say, however, that
the effect of the vapor generator loop was to eliminate hydrocarbon
emissions under conditions where there were some measurable
emissions. This effect is undoubtedly attributable to increased
residence time and gradual quenching.
3.	Influence of Non-Uniform Flow Distribution
on Emissions
During the course of the experimental work
performed here, a number of modifications were made in the test set
up and instrumentation as it became evident that a problem
existed in one or the other of these. The problems involved in the
measurements of emissions are discussed elsewhere. There was,
however, a problem involved in the hydrocarbon emissions which
sheds some interesting light on the problems of emission control.
The early vapor generator exhaust data showed large
quantities of hydrocarbons in the vapor generator exhaust which
were not found in measurements made directly from the burner.
This peculiar behavior suggesting that a source of hydrocarbons
existed between the burner and the top of the exhaust
stack was investigated extensively.
An examination of the flame from the burner inlet
showed that there was a substantial maldistribution of fuel in
the inlet stream, causing rich mixtures to occur in some portions
of the burner with lean mixtures in the other portions. When
the improper fuel injection pattern was modified to provide
a uniform mixture, the anamolous hydrocarbon reading in the
vapor generator exhaust immediately disappeared. Correction of the
fuel infection pattern was made between runs 279 and 282.
The maldistribution of flow from the fuel injector
influenced all of the emission levels, not just that of hydrocarbon
data. Examination of NOx data from the burner after the fuel
injector was corrected show a reduction in NOx as a
function of mixture ratio. This suggests that the local mixture
ratio and local temperature are the principal factors in the
formation of oxides of nitrogen, and that the uniform mixture
which yields the lowest maximum temperature in the burner also
provides the lowest overall emission levels.
4.	Theoretical Analyses
111-6

-------
Theoretical analysis of the burner performance
has been conducted using the heat balance theory described
by Vulis, Longwell and others. The theory permits an analysis
of several interesting parameters in the burner operation. These
include burner stability limits, and the prediction of burner
emissions in the form of unburnt material as a function of burner
design and operating conditions. The theoretical analysis was
conducted to determine the influence of the burner volume, air
flow rate, ambient pressure and inlet temperature on burner
stability, burner efficiency and combustion temperature. Only
lean burning cases were treated • Prediction of lean blowout
limits and burner emissions due to incomplete combustion were
made for each lean operating point for comparsion with the
experimental data.
An interesting feature of the analysis is the prediction
that at elevated inlet temperatures, on the order of 1600°F and
higher, the burner will no longer show critical burner character-
istics at very lean mixture ratios. Thus at high inlet
temperature and low combustion temperature it appears that the
burner will ignite and burn independently of the air flow. Under
these conditions the burner will not exhibit blowout character-
istics, but the efficiency of combustion will still be strongly
dependent on the flow rate and other factors.
5. Data Correlation
The theoretical analyses conducted as a part of
this program were used to assist in the correlation of experimental
data from the Paxve burner.
a.	Correlation of Stability Data
Stability data from the Paxve Burner
shows very good correlation with the blowout predictions based on
the burner theory. At very low flow rates, the burner seems to be
somewhat less stable than the theory would predict. This lessened
stability has not been examined in detail but is probably
attributable to heat loss from the burner. The theory appears to
account properly for the influence of air flow, inlet temperature
and burner volume.
b.	Correlation of Carbon Monoxide and
Hydrocarbons Emissions Data
The carbon monoxides emission data has
been examined in the light of the combustion theory outlined in
this report. What we expect is that as we approach the lean
stability limit, the efficiency will decrease, and this
inefficiency will show up as either unburned hydrocarbons, carbon
monoxide, or both. Experimentally, we find that with propane
combusiton there are virtually no unburned hydrocarbons in the
exhaust as the lean limit is approached, but there is a relatively
large amount of carbon monoxide.
The shape of the carbon monoxide emission curves is
III-7

-------
substantially the same as that predicted by the theoretical
analysis. The influence of the flow rate and temperature on
the carbon monoxide emissions is also in the direction we would
expect based on the experimental data and the theoretical
analysis. This is probably due to the simplifications used in the
burner analysis. Heat release associated with partial combustion
of the fuel to water vapor and carbon monoxide was ignored. This
tends to underestimate the efficiency as lean blowout is approached,
and to overestimate the emissions. A more exact expression for the
heat release could probably yield still closer agreement with
experimental data. The present analysis is useful, however, for
obtaining a conservative estimate of burner CO emissions.
The influence of flow rate, inlet temperature and mixture
ratio on the hydrocarbon emissions near lean blowout with kerosene,
shows similar characteristics to those obtained for the carbon
monoxide. The effect of fuel/air ratio, flow rate, and inlet
temperature are along the lines predicted by the theory.
The emissions of both CO and HC are greatly diminished by
passage through the vapor generator. This is probably due in part
to continued oxidation as gases cool off. The lower CO values
at fuel/air ratios approaching stoichiometric is also attributable
in part to shifting equilibrium in the flow through the heat
exchanger.
We have not attempted to correlate the reductions in
CO arising from passage through the vapor generator. By using the
predicted unreactedness to estimate the CO, we thus have an even
more conservative estimate for a burner/heat exchanger installation.
c. Correlation of the NOx Data
We do not have a theoretical basis for
correlating the oxides of nitrogen data which is totally in agree-
ment with the experimentally determined values. It appears that the
oxides of nitrogen data can be correlated as a function of the
combustion temperature. If one writes an equation of the form:
[NO] = Ke"E/RT
and use this as a basis for correlation of the experimental data
we find that:
K = 4.38 x 10s ppm
E = 36.8 K cal/mole
seems to give a fair fit to the data.
The empirical correlation produced here is of value
since it permits a relatively accurate prediction of the oxides
of nitrogen content in the burner exhaust as a function of operating
conditions.
III-8

-------
6. Emission Measurement Problems
A number of emission measurement problems were
covered during the course of the experimental investigation reported
here. The most important of these dealt with erroneous measurements
which can be made in the measurement of oxides of nitrogen due
to improper technique. It appears that some of these improper
techniques are widespread. Additional discoveries deal with the
problems of measuring hydrocarbons at very low levels.
a.	Oxides of Nitrogen Measurement
Problems
Difficulties experienced in measurement
of oxides of nitrogen by the Griess-Saltzman method during the course
of this program are discussed in some detail in this report. These
problems included saturation of the dye caused by using too
small a quantity of reagent for the volume of the gas being sampled,
and failure to detect NO in rich mixtures due to the lack of oxygen
in the flask to oxidize the NO into NO2. Those errors on our part
were avoidable. Someone who was more familiar with Saltzman's work
and with the standard measurement procedures used in the automotive
field would probably not have experienced those difficulties.
An additional problem was discovered, however, which to
the best of our knowledge is not discussed elsewhere. At low levels
of concentration in the exhaust from our burner, N02 appears to form
a substantial portion of the total NOx. It further appears that
the NO2 which is highly soluble in water, can be readily lost
prior to analysis by improper sampling. The use of a sampling
procedure which permits or encourages the water formed during
the combustion process to be removed from the flow before analysis,
can be expected to give erroneous results at low NOx concentrations
due to the loss of NO2 . It is probable that other forms of
water removal equipment, such as desiccants and absorbers will also
tend to trap NO2.
Removal of water vapor from the exhaust stream prior to
analysis is a common procedure in NOx emissions measurements. In the
case of NDIR analyzers which have been widely used in the past, for
NO measurement, water vapor must be removed because it interferes
with the detection of the NO. It is our conclusion, based on the
work done at Paxve, that every effort should be made to eliminate
water dropout if all of the oxides of nitrogen are to be
detected in a low NOx stream.
b.	Problems in the Detection of
Hydrocarbon Emissions
Difficulties experienced at Paxve in
the detection of hydrocarbon emissions were of three types. These
included the oxygen synergism effect on the flame ionization
detector, the problem of inaccurate measurement due to use
of a cold sampling line, and the problem of zero shift of the
instrument, apparently due to water vapor in the exhaust.
III-9

-------
Oxygen synergism is a name given to the change in
sensitivity of the flame ionization detector to hydrocarbon in the
presence of oxygen. The change in sensitivity depends on both the
oxygen concentration and on the hydrocarbon being detected. For
hydrocarbons of interest to the analysis being conducted by Paxve
this effect was on the order of 25% reduction in sensitivity at
10% oxygen concentration. The oxygen synergism effect could
have been minimized by the use of a nitrogen hydrogen mixture
as the combustible gas in the flame. The manufacturer who
supplied the flame ionization detector which was used did not
advise us in this regard. We discovered the problem during
instrument calibrations.
Heating of the hydrocarbon sampling line and pumping
equipment is necessary to obtain accurate measurement of hydrocarbon
emissions. There seems to be some misconception as to why
this heating is desirable. It was suggested to us that heating
of the lines was an effective means for preventing condensation
of higher hydrocarbons. While this is true, the concentration of
hydrocarbons necessary to actually permit condensation even at
ambient temperatures is quite high. That concentration is
unlikely to occur in the exhaust from a clean burner such as
the Paxve burner. The problem which does arise in this regard is
not one of true condensation, but rather psuedo-condensation
caused by the adsorption of the hydrocarbons in the surface of
the sampling line. Heating of the sampling line and pump
undoubtedly serves to eliminate adsorption of higher hydrocarbon
in the same fashion.
Perhaps the most vexing problem which arose in the
measurement of hydrocarbons during the course of this program,
was the repeated observation of negative values of hydrocarbon
emissions from the burner and the vapor generator exhaust. Negative
readings of as much as 0.5 ppm expressed as hexane were common
place. Efforts to attribute this to a drift in the instrument
were in vain. Preliminary experiments with zero air into which
10% water vapor was evaporated showed a zero shift of about 0.5 ppm.
This was a zero shift as opposed to a change m sensitivity of the
sort associated with oxygen synergism.
Thiz zero shift presents somewhat of a dilemma. On the
one hand, we would like to remove the water vapor from the flow
so as to avoid the zero shift. On the other hand, all methods
for eliminating the water vapor from the flow that we have been
able to think of would simultaneously influence the measurement
of the hydrocarbon content. It appears that the only practical
method for taking this into account is to measure the water
vapor content and use this together with known calibrations of
zero shift to correct the data.
We used pure hydrogen in our FID. We do not know whether
the zero shift would be as pronounced with some other gas mixture.
A 40/60 hydrogen-helium mixture is commonly used in FIDs to minimize
the oxygen synergism effect. This might also influence the zero
shift.
111-10

-------
c. Carbon Monoxide Measurement Problems
Most of the CO data from the vapor
generator exhaust showed levels which were well below the detection
limit of our equipment. We were delighted and somewhat surprised
with the low CO emission levels from our burner. It would have
been nice to be able to measure them more accurately. The equipment
which we used permitted us to make carbon monoxide measurements
down to about 5 ppm. Obtaining lower readings through ordinary
gas chromatography is quite difficult since the carbon monoxide
separation from nitrogen on the molecular sieve is not as complete
as one might like. More sensitive techniques for measurement of
carbon monoxide have been devised. These include a method for
converting carbon monoxide into methane and then measuring the
methane in a flame ionization detector, and highly sensitive NDXR
instruments.
7. Summary
In its normal mode of operation, the Paxve
combustion process is capable of very low emission operation.
Oxides of nitrogen on the order of 0.1 gm/Kg are achieved simultan-
eously with carbon monoxide levels of less than 1.0 gm/Kg and
hydrocarbon readings which are so low as to be undetectable.
The Paxve combustion process appears to be one which
requires more sophisticated measurement devices than those currently
available for testing vehicle emissions.
B. Recommendations
1.	Prototype Burner Development
A program should be funded to support the
development of the Paxve burner into a prototype unit suitable
for incorporation into a Rankine Cycle Engine.
a.	Burner
A prototype burner vapor generator
assembly would be a desirable line of approach.
b.	Fuel Vaporizer
In conjunction with this type of
prototype development, a fuel vaporizer development program should
also be funded. The fuel vaporizer which was used in conjunction
with the kerosene burner data reported here, should serve as the
basis for development of a practical prototype.
2.	Liquid Injection Investigation
Further reasearch should be conducted into
the factors affecting the emissions characteristics of the Paxve
burner. Limited data on liquid kerosene injection shows that under
some conditions the burner seemed to operate with low emissions
III-ll

-------
using very poorly atomizing liquid infection. This should be
explored further to determine whether fuel vaporization is a
necessary feature in the development of this burner. This program
should not take precedence however, over the prototype development
suggested as recommendation 1 above. It is important that practical
application of this promising low emission technology be made,
with product improvement of somewhat less priority.
3.	Transient Effects on the Paxve Burner
Further research on the Paxve burner is desirable
to investigate the effect of transient operation on burner
emissions. Transient changes in flow rate through the burner were
not examined during the course of the program discussed here.
Such measurements would be highly desirable. This work might
proceed in parallel with the prototype development suggested above.
4.	Influence of Vapor Generator
Additional research on the influence of the
vapor generator on the Paxve burner emissions, particularly on the
carbon monoxide and hydrocarbon emissions would be desirable.
Continuous measurement equipment is mandatory if this type of
development supporting research is to be conducted in an
efficient manner. The grab sample techniques utilized by Paxve
during the course of the program described here were selected
because of their availability and accuracy. The advent of the
chemiluminescent NOx technique and high sensitivity NDIR for
carbon monoxide should permit continuous flow instrumentation for
these parameters. Continuous flow instrumentation for the
hydrocarbons is already available and was found to be quite
adequate for the program.
Carbon monoxide and hydrocarbon emissions improve
greatly in passage through the vapor generator under lean
operating conditions. It would be desirable to determine the extent
to which this is the case, since it is the balance between the
carbon monoxide and the oxides of nitrogen emissions which limit
the operating range of the burner otherwise.
5.	Elevated Pressure and Temperature Effects
Further research into the influence of pressure
and temperature on the Paxve burner should be funded. The Paxve
burner may be applicable not only to Rankine cycle engines, but
also other forms of combustion equipment, including gas turbine
engines. If a Paxve burner is to be used as an element in an
automotive or stationary gas turbine, it will have to operate
under conditions of elevated inlet pressure and temperature which
are considerably outside the range of the testing accomplished
to date.
The influence of elevated pressure and temperature are
to some extent predicted by the theoretical analysis conducted here.
Only experimental investigation of these phenomena will verify the
validity of that analysis . A preliminary examination of this
111-12

-------
problem suggests that full scale testing of burners of a size
suitable for gas turbine application would be the most feasible
approach to conducting research in this area.
6. Extended Burner Analysis Including Heat Loss
and NOx Correlation
Additional analysis should be funded to extend
the theoretical investigations discussed here and to improve the
experimental correlation with theory.
a.	Burner Theory
The burner theory discussed here
showed excellent agreement with the experimental blowout data at
high flow rates. At lower flows there was some deviation. Improve-
ment in the theory to account for heat loss from the burner would
undoubtedly resolve this discrepancy. It would also show whether
or not other types of burners which are more susceptible to heat
loss can be expected to achieve the same excellent results that
we have achieved with the Paxve burner.
The burner heat release equations should also be
modified to account for some heat release when partial oxidation
of the fuel to water vapor and carbon monoxide takes place. This
should improve the accuracy of the CO emission predictions.
b.	HOx Correlation
Additional correlation of the oxides
of nitrogen data obtained during this program would be highly
desirable. The fact that the oxides of nitrogen do not appear
to be strongly influenced by the air flow rate suggests that
combustion related atomic oxygen may be a significant factor.
This should be investigated analytically.
111-13

-------
IV. EXPERIMENTAL METHODS
This section covers a description of the experimental
apparatus, critical instrumentation, and test procedures used in
obtaining the experimental data. Special test facilities were
designed and fabricated to permit testing the Paxve burner under
controlled conditions of inlet fuel and air temperature and pressure
for the evaluation of emissions and stability.
A. Experimental Installations
1. General Description
All of the burner emission and stability testing
reported here was accomplished in the Paxve combustion laboratories.
These include two primary combustion test facilities: Stand 1 and
Stand 2 ("The Blockhouse"). A schematic diagram of the overall
burner test facilities incorporating burner test stands No. 1 and 2
as well as the auxiliary instrumentation and gas analysis facilities
is shown in Figure 1. The two burner test stands are nearly
equivalent in terms of burner testing capacity, instrumentation, and
test fuels. The major difference is that test stand 2 contains a
working fluid vapor generator loop. This allows study of the effect
of burner gas quenching on emissions.
As indicated in the diagram of Figure 1, the two test
stands are self contained for individual burner testing. They
consist of a burner control console, an enclosure in which the burner
is installed, an air system which supplies combustion air to the
burner, a fuel system supplying the particular fuel under test, an
instrumentation system which provides direct readout for control of
burner operating conditions, and instrumentation for measurement of
the burner exhaust gas emissions. External view photographs of
burner test stand No. 1 and control console for test stand 2 are
shown in Figure 2.
Test stand No. 2 is constructed to perform the same
functions as burner test stand No. 1 with the additional feature
that a vapor generator system has been added to the burner. The
burner and vapor generator systems are totally enclosed in a well
ventilated blockhouse for safety reasons. Photographs of burner
test stand No. 2 showing the interior view of the blockhouse are
shown in Figure 3.
The two burner test facilities are served by a common gas
sampling instrumentation center. This instrumentation center
contains the instrumentation for the measurement of exhaust gas
emissions which include CO, CO2, unburned hydrocarbons, and oxygen.
In addition, gas analysis equipment is provided for the
colorimetric evaluation of oxides of nitrogen in the exhaust. The
instrumentation and emissions measurement techniques are described
in detail later in this report. Photographs of the emissions
instrumentation center are presented in Figure 4.

-------
2. Air Flow Subsystems
A schematic diagram of the air and fuel system
for burner stand No. 1 is shown in Figure No. 5. The schematic
diagram for Stand No. 2, is shown in Figure 6. The air system
consists of two variable speed air blowers connected in parallel,
capable of providing up to 250 lb/hr of air. The blowers are
controlled by a variac to set blower speed and consequently air
mass flow. Air flow is measured through a rotameter. The air
temperature is varied by means of a nichrome wire air heater
immersed in the air flow. Air temperatures of up to 500°F can be
obtained. Current flow in the nichrome wire is regulated by a
variac auto transformer. The thermally conditioned air flow is
discharged coaxially with the injected fuel flow into the burner
for ignition and combustion.
3. Fuel Flow Subsystems
The fuel systems consist of two parallel storage
and control systems, one to handle propane, the other for diesel/
kerosene fuels. A major element of the fuel system is the fuel
heater vaporizer. The heater/vaporizer for Stand 2 is shown in
the photograph of Figure 7. Figure 8 shows a drawing of the unit
which was designed and fabricated to meet the specific requirements
of this burner evaluation program. It consists of a high thermal
conductivity core section containing an encapsulated electrical
heating element. Helical grooves are machined on the outside of
the core which provide a vaporization passage for the fuel flow.
The encapsulation of the electrical heating element in a high
thermal conductivity aluminum core mass provides isothermal heat
transfer at the helical fuel heating passages. Maintaining
uniform wall temperatures through mixed phase flow eliminates "hot
spots" which could lead to thermal decomposition (carburization)
of the hydrocarbon fuels.
The outside case of the heater consists of a tube with a
side wall inlet to the spiral fuel grooves and an outlet through
a domed end piece. During operation, the fuel enters through the
case and travels up the spiral path to the dome and cap. The end
cap is designed so that liquid vapor separation is achieved with the
vapor leaving through a centrally located port. Unvaporized
droplets are retained by the dome to fall back onto the heated core
for further vaporization. The unit is thermostatically controlled to
achieve the desired vaporization for various fuel flow rates.
Figure 7 shows the disassembled vaporizer after operation
on kerosene. More than 20 gallons of kerosene and 1000 lbs of
propane have flowed through the heater unit. The core operated at
approximately 900°F at a flow rate of approximately 7 lb/hr.
Examination of the vaporizer shows only a thin powdery layer of
black carbon covered the top of the core where the flow could
stagnate. The helical flow passages discolored but were free from
any carbon deposit.
As shown in Figure 6 a line heater is also installed
between the fuel vaporizer and the burner fuel injector to maintain
IV-2

-------
fuel temperature at the desired value after vaporization. The
line heater consists of a resistance heated length of quarter inch
stainless steel tubing. Electrical control is achieved through
a current transformer and control Variac. The line is insulated to
minimize the convection losses.
This fuel conditioning system has demonstrated good
performance. Vaporized fuel or hot propane can be delivered to
the burner at any desired temperature up to 800°F over a flow
range from less than 1 lb/hr to more than 12 lb/hr.
4. Vapor Generator System
The system which is unique to Stand 2 is the
vapor generator system, a schematic of which is shown in Figure 9.
The purpose of this installation is to study the effects of
quenching by cold surfaces on burner emissions. The vapor generator
itself is a counter flow heat exchanger in which the working fluid
circulates through a 6 foot tall coil of 3/8 in. O.D. tubing. The
coil dimensions after winding were 2 in. I.D. x 2 5/8 in. O.D.
The coil is surrounded on the outside by 3 in. I.D. ceramic tubing
and is bordered on the inside by a 1.5 in. O.D. air tube leading to
the burner. The working fluid runs counter flow in the coil to the
hot burner gases which rise in the annular area between the center
air tube and the ceramic wall. Figure 3a is a photograph of the
vapor generator stack. Figure 10 shows a drawing of the unit.
The working fluid is circulated by a geroter pump which
is driven by an air cooled Volkswagon Automotive engine. The pump
can provide outlet pressures of up to 1500 psi at flow rates of
7.8 gpm at 120 0 rpm. The automotive drive engine is provided with
remote control ignition, starting, throttling and clutching
capability to vary pump speed and circulating flow rate.
The initial design of the vapor generator loop included a
jet pump at the inlet to the gerotor pump to prevent cavitation. It
was found that this did not permit the system to start flowing
properly. The jet pump must have a supply of high pressure fluid
in order to provide the desired characteristics. For our case a
startup problem results since the pump cannot supply high pressure
unless it has full flow at the inlet. In a future automotive
system, this difficulty could be overcome by the slow cranking of
the pump during initial system starting. It was beyond the scope of
the present test facility installation to provide this necessary
flexibility, and the jet pump was removed. The cavitation problem
was circumvented by installing a low pressure accumulator
pressurized with N2 at the pump inlet to set the minimum pressure
level in the system. This provided adequate cavitation supression
and also a reserve fluid supply.
The system also includes a high pressure accumulator at
the pump outlet to minimize water hammer effects. For additional
flow control, a pump bypass line is provided which is manually
operated and set at a given condition which is normally unchanged
during testing. Provisions are made for filling the system at the
low and high pressure accumulators while bleeding gas at a high
IV-3

-------
point bleed. A filter at the pump inlet is provided to protect the
close tolerance of the pump rotor.
The vapor generator system has been operated in conjunc-
tion with the burner to working fluid outlet temperatures of 500°F
and flow rates of 2.0 GPM through the vapor generator coil. The
vapor generator system also includes a condenser coil which is
cooled in a water bath and returns the working fluid temperature
to approximately 212°F before entering the pump.
5. Control Consoles
Each test stand is provided with a control
console which incorporates all necessary operating controls and
visual monitoring instrumentation. The operating controls
include: switches for the blowers, heaters, igniter, and fuel flow
solenoids; Variacs (autotransformers) to control air flow rate
and electrical power to the air heater and the fuel line heaters;
needle valves for control of fuel flow and nitrogen pressurizing
and purge flows; and a thermostatic controller for the fuel heater/
vaporizer. Instrumentation includes: flow meters (rotameters) for
fuel and air flow rate; pressure gages for fuel supply and delivery,
burner air supply pressure, pressure at the rotameters, and
pressure levels throughout the vapor generator loop; and tempera-
tures in the burner, at the vapor generator coil exhaust, in the
fuel and air streams both at the flow meters and at the burner
inlet, and throughout the vapor generator loop. A table of all
control and measurement functions on the two operating consoles is
presented in Tables 1 and 2.
B. Instrumentation
The burner test stands are provided with pressure,
temperature and flow instrumentation. These instruments permit the
test operator to set and hold a desired test condition. The
instruments together with the corresponding controls (described
in A-5 above) are tabulated in Tables 1 and 2.
In addition to the test operation instruments, Paxve has
provided emissions measuring equipment which incorporates measure-
ments for CO, CO2, O2, HC, and NOx.
1. Pressure Measurements
All pressure measurements involved in the burner
research program were made with conventional Bourdon tube or
diaphragm (Magnehelic) pressure gages. Pressure measurements were
made for several purposes:
a.	The pressure level of each fuel tank was
monitored to help set up repeatable test points.
b.	The pressure level at the propane flow
meters was used to correct propane flow rate for density.
c.	Pressure drop across the fuel vaporizer
IV-4

-------
was monitored to see if the vaporizer was loading up with carbon
deposits.
d.	Pressure levels throughout the working fluid
loop were monitored to ensure that the working fluid was above its
critical pressure in the vapor generator and also to avoid cavita-
tion problems in the gerotor pump. Pressure drop across the filter
was also monitored.
e.	Pressure loss through the burner was
measured.
f.	Pressure level at the FID input was
measured to control the flow rate through that instrument.
2. Temperature Measurements
All temperature measurements were made with bare
junction thermocouples and microammeter type readouts (Assembly
Products, Inc. "Simplytrol" Meters). Iron-constantan and chromel-
alumel couples were used for low temperatures. Platinum vs
platinum-13% rhodium couples were used to measure the burner
temperature.
Temperature measurements were made for the following
purposes:
a.	The air temperature was measured at the air flow
meter to permit correction of flow rate for air density. Although
the air flow was measured before the air was heated, there was
significant heating of the air by the air blowers.
b.	The fuel temperature at the flow meter was measured
to permit density correction.
c.	Air temperature was measured downstream of the air
heater.
d.	Fuel temperature was measured downstream of the
fuel/vaporizer and also downstream of the heated fuel line.
e.	The temperature of the fuel heater/vaporizer was
measured by a thermocouple buried in the aluminum core. The
thermocouple was connected to an automatic controller which
kept the core temperature within a few degrees of a set point.
f.	The temperature in the burner was measured by a
ceramic encased Pt-Pt 13% Rh thermocouple inserted through the
walls of the burner. Early tests in Stand No. 1 used a movable
probe which could be inserted into the burner through the air
intake pipe. It was found that this sometimes interfered with
the normal combustion process and was discontinued. Measurements
of the gas temperature in the burner exhaust was substantially
the same as the measurement made inside the burner. Burner
temperature measurements were made primarily to tell if the burner
was lit and/or operating stably. Flame out could be detected
IV-5

-------
almost immediately, but incipient flame out was difficult to detect.
Because the burner temperature probe was not shielded, it suffered
from inaccuracy due to radiant heat loss. This generally resulted
in a 200°F to 300°F difference between the burner temperature
reading and the theoretical flame temperature.
g.	The temperature of the burner exhaust gases was
measured downstream of the vapor generator stack.
h.	The temperature of the fluid within the vapor
generator loop was measured at several points including upstream
and downstream of the "condenser".
i.	The temperature of the pump and line used for
hydrocarbon sampling were monitored to be sure that these
components were above 300°F.
3. Flow Measurements
Flow measurements for the experimental program
reported here were made with the variable area flow meters commonly
referred to as"rotameters". All flow meters were calibrated
throughout their range at ambient pressure and temperature, with
appropriate corrections for gas density as necessary. The
propane flow meters were also calibrated at elevated pressures
to verify the density correction technique.
Flow measurements included:
a.	Air flow to the burner, increased upstream of the
air heater. Air temperature at the flow meter was measured for
density correction.
b.	Fuel flow into the burner, measured upstream of
the fuel heater/vaporizer. Both high and low range rotameters
were used for propane flow. Pressure and temperature at the flow
meter exit were measured for density correction. One wide range
flow meter was sufficient for kerosene flow measurements on
Stand 2. Kerosene flow on Stand 1 was measured with a low flow
rotameter and one of the propane meters, recalibrated for the
liquid fuel.
c.	Flow of working fluid through the vapor generator
loop, measured upstream of the pump inlet. The temperature of
the liquid at the flowmeter was measured to allow density
correction.
d.	Flow of sample gas through the Flame Ionization
Detector Hydrocarbon Analyzer. This was not measured directly,
but was maintained at a constant value by heating-any gas flow
through the instrument to the temperature of the sample line, and
by holding a fixed pressure ahead of the capillary line restriction
at the instrument inlet.
e.	Flow of helijUm through the thermal conductivity,
detector gas chromatograph. This was measured by using a "soap
IV-6

-------
bubble flow meter". The helium flow rate was set to 15 cc/min each
time the instrument was used.
4. Emissions Measurements
Measurements of the burner exhaust gas were
made for oxides of nitrogen (NOx), carbon monoxide (CO), unburned
hydrocarbons (HC), carbon dioxide (CO2) and oxygen (02)- The
instruments and techniques used are described in detail in section
V of this report. Briefly they were:
a.	Griess-Saltzman method for NOx. This is a
wet chemistry method that requires drawing a sample into a
container that contains a chemical solution which turns a reddish
purple color in the presence of NO2.
b.	Thermal conductivity detector gas chroma-
tograph for CO and CO2. This instrument permitted analysis of
discrete samples of burner exhaust which were supplied by a sampling
pump.
c.	Flame Ionization for hydrocarbons. This
instrument permitted continuous analysis for total hydrocarbons. It
required a heated sampling line and heated sample pump. The FID is
part of a chromatograph which can be used for hydrocarbon
separations and analysis.
d.	Bailey Heat Prover for O2 and combustibles
(CO and H2). This instrument draws a sample from the burner flow
providing continuous analysis. This instrument was the principal
means used for setting the burner fuel/air mixture ratio. It
provides an oxygen reading which was in close aqreement with the
more accurate values obtained by volumetric analysis.
3. Volumetric gas analyzer for CO2, 02, and CO.
This instrument, commonly referred to as an "Orsat" apparatus was in
fact manufactured by the Burrell Corporation. It provides an
accurate analysis of a gas sample drawn into the instrument.
Volumetric analysis is obtained for CO2, 02, and CO to an accuracy of
about 0.1%.
C. Test Procedures and Techniques
Testing was conducted for the general purpose of
investigating and documenting the emissions characteristics and the
stability characteristics (rich and lean blowout limits), of the
Paxve burner. Testing was conducted in burner test stands 1 and 2.
Test stand 1 was used primarily for stability testing and burner
emissions with propane. Stand 2 was utilized for the detailed
investigation of the influence of the vapor generator system on
burner emissions and for kerosene stability testing.
1. Burner Operation
The general burner test procedure was to ignite
the burner and bring it to the desired combination of air mass flow,
IV-7

-------
air and fuel inlet temperature and fuel/air ratio. Ignition of the
burner with propane was accomplished by setting a low airflow,
turning on the igniter and bringing in the fuel so as to approach
ignition from the lean ignition limit. Ignition in this case was
always smooth and continuous. Once the burner was ignited, air flow
and fuel flow were increased simultaneously to the desired total
mass flow and fuel/air ratio setting. In the case of kerosene
combustion a burner warmup period was usually allowed during which
the burner was ignited and operated with propane. Then after
the burner and lines were hot the kerosene was brought in and
propane flow reduced until kerosene combustion was self sustaining.
The burner test condition was established by visual
readout of the console pressure, temperature, and flow meters, and
the Bailey Heat Prover. The latter instrument gives continuous
reading of the percent oxygen and percent combustibles in the
exhaust gas. The combustion data of Figure 11 shows the variation
of the percent oxygen and combustibles with fuel/air ratio for
operation with propane and octane. It should be noted that the
combustibles reading on the Bailey is a combination of the CO and
H2 content in the gas.
Once a stablized operation test point is established
readings are taken of all the operational instruments. Chromato-
graph records are taken and identified on the strip chart. In the
case of the 9000 instrument (FID), the continuous level of total
hydrocarbons is identified with a run number. The recorder is then
switched to the left channel of 8004 instrument to measure carbon
monoxide and oxygen using the left hand column. The right channel
and column are then used to measure carbon dioxide. Burner gas
samples are collected in flasks for Greiss-Saltzman analysis of
N02- Concurrently, a gas sample is drawn into the Burell volumetric
analyzer and is then analyzed for CO2, oxygen, and CO. After all
data has been recorded and spot checks made to establish consis-
tency, a new operational test point is established by adjusting the
air flow, or an inlet temperature.
Testing of the burner was conducted over a wide range of
fuel/air ratios, ranging from the lean limit to the rich limit of
burner operation. At times these tests were destructive to the
burner or portions of test apparatus. This was particularly true at
stoichiometric and rich burning operation. Stoichiometric combus-
tion of propane or kerosene with air leads to combustion tempera-
tures on the order of 3500°F to 4000°F. * Such temperatures are well
m excess of the structural capabilities of any of the materials
used for the fabrication of the Paxve burner.
Fortunately, the burner materials do not reach the
theoretical flame temperature, and therefore the burner is
capable of surviving periods of stoichiometric operation. The
ceramic portions of the Paxve burner can withstand temperatures
on the order of 3300°F, but the metallic portions of the structure
do not have these capabilities and hence destruction of various
metallic sections such as inlet pipes or external supports
sometimes occurred.
IV-8

-------
When rich combustion was being investigated, there were
in general two combustion processes going on. Combustion within
the burner was taking place using the burner air supplied through
the inlet pipe. This internal combustion during rich operation
might involve combustion temperatures on the order of 2500°F to
3000°F, well within the burner capabilities. Simultaneously,
however, after-burning was taking place outside the burner where the
burner exhaust mixed with ambient air. That external combustion
process impinged directly on some portions of the test stand or
inlet tubing. Protection of these elements was therefore necessary.
Potential destruction of the test stand sometimes limited the
capability for extensive stoichiometric or rich testing.
Attempts were made early in the program to make measure-
ments of the combustion air and burner temperatures by means of
probes extending through the air inlet pipe. It was found that
these probes sometime interfered with the operation of the burner
causing flash back or flame holding on the probe. This was in
general accompanied by rapid deterioration of the inlet pipe. Inlet
pipe probing was discontinued in order to avoid this type of
problem.
During some tests near stoichiometric operating
conditions, combustion within the burner inlet pipe would occur,
particularly at low flows with high inlet temperatures. Data were
gathered under these difficult conditions to the fullest extent
possible. This type of phenomenon did, however, restrict the
range of burner operating conditions we could conveniently explore.
2. Vapor Loop Operation
For those tests in which measurements were made
downstream of the working fluid vapor generator, it was necessary
to start the working fluid system before igniting the burner. This
avoided problems of overheating the fluid in the coils which could
otherwise have taken place.
Once the vapor loop system was circulating working fluid
at an acceptable rate, the burner was ignited and the usual
procedures previously described for burner operation and data
measurement were followed.
Start up of the vapor generating loop was initiated by
pressurizing the low pressure reservoir to about 30 psi. This
eliminated problems of pump cavitation during start up and operation
of the system. The pump drive was then started and the loop brought
up to an acceptable operating condition. Early in the program, a
remotely actuated clutch between the engine and the working fluid
pump was disengaged when starting the motor. The clutch was then
engaged in order to drive the pump. It was found that this
disengagement and re-engagement of the clutch was unnecessary and
in later stages of the program the clutch was left permanently
engaged.
Operation of the vapor loop was generally uneventful
unless the nitrogen gas bubble at the top of the high pressure
IV-9

-------
accumulator had been lost. When there was no gas bubble in the high
pressure accumulator, the pump pulsations were fed to the pressure
gauges resulting in physical damage to those gauges.
All of the operating parameters of the vapor generator
system were monitored including the working fluid flow rate and
the liquid and burner exhaust inlet and exit temperatures. No funds
were provided m the contract to investigate the performance of the
vapor generator from an efficiency point of view and therefore this
data has not been reduced.
Exposure of the working fluid to excessively high
temperatures in the presence of even small amounts of oxygen causes
thermal degradation of the working fluid. The nitrogen utilized for
the low pressure reservoir pressurization system and the high
pressure accumulator contained small but nevertheless significant
quantities of oxygen. Care was taken, therefore, to avoid
excessive vapor generator tube wall temperatures. Vapor outlet
temperatures as high as 600°F were occasionally permitted, but
in general the vapor loop exit temperature was kept below 400°F.
Since the purpose of the vapor loop was to quench reactions in the
burner exhaust it was not felt necessary to permit higher working
fluid temperatures.
Operation of the burner in connection with the vapor
generator loop was restricted to a fairly narrow range of burner
operating conditions as a further step towards avoiding problems
with the working fluid. The burner itself will operate from an
oxygen concentration near the lean limit of about 12% through
stoichiometric to a combustible concentration at the rich limit
of over 20%. Unfortunately the operation of the burner at
oxygen concentrations of less than about 6% to 8% yields gas
temperatures which are too hot for safe operation of the vapor
generator loop.
The policy, therefore, was to operate the burner in
connection with the vapor generator from its lean limit up to
a maximum combustion temperature of about 2 700°F. This corresponds
with ambient inlet temperature to about 6% oxygen concentration.
With an elevated inlet temperature (T air 400°F) , the oxygen
concentration in the burner exhaust was not allowed to
fall below 8%.
Operation of the vapor generator in connection with the
burner required special system shut down procedures. The burner was
shut off by cutting the fuel flow with the air blower still
operating, the blower was then continued in operation until the gas
temperature level leaving the burner was below 500°F. Working
fluid circulation was maintained during this time. This precaution
during shutdown was necessary to avoid overheating the working
fluid in the coils.
Instrumentation for monitoring the burner operating point,
specifically the Bailey Heat Prover and the volumetric gas analysis
equipment, were set up to draw their gas samples from the exit to
the burner, ahead of the vapor generator loop. This permitted rapid
IV-10

-------
response of the system to changes in burner input variables. It
also avoided questions of inadvertent dilution of the burner exhaust
with ambient air which occasionally arose when the system was
improperly set up.
The exit temperature of the gases from the vapor generator
exhaust were monitored as well as the temperature of the gases in
the burner. Vapor generator air exhaust temperature was frequently
in the range of 200°F or less. During start up it was common to see
water condensing on the vapor generator coils. This condition
persisted until the working fluid became hot enough to cause
evaporation. It was found that the presence of water on the vapor
generator coil and air intake tube surfaces influenced the oxides
of nitrogen readings from the top of the vapor generator stack.
This influence was duly noted during the course of the program.
3. Blowout Test Procedure
Lean and rich blowout limits are defined as
the demarcation between the burner's ability to burn or not burn
due to a change in fuel or air flow. It is not actually possible
to operate at the blowout limit. The blowout limits are found by
showing that a burner will stay lit at a particular set of
conditions (a specific fuel/air ratio) but not stay lit at another
set of conditions (another fuel/air ration) not far removed from the
first set of conditions. Further, it is not sufficient that the
burner merely operate for a short period of time at a set of
conditions, but is required that sustained (steady state)
operation be evident so that an equilibrium condition exists
within the burner.
Determining blowout conditions, then, requires that the
burner operate for a sustained period of time while being observed
to see whether the operation is steady or if the burner is going
out. Determining blowout limits with propane or kerosene involves
substantially the same operation, so that a description of one will
characterize the other. The determination of lean limits is
somewhat different than determining rich limits. Both methods will
be described.
The emission measurement equipment needed for the tests
are (1) the Bailey Heat Prover, (2) the FID hydrocarbon detector, and
(3) the volumetric gas analyzer. The gas chromatograph (thermal
conductivity detector) may be used, but it was not essential for
these tests.
a. Lean Blowout Using Propane
A burner is set up and ignited, as
previously described in the burner operation procedure. The air
flow rate is set at the desired value. The fuel flow rate is set
to obtain a reasonably hot burner (f/a = 0.04 to 0.05) to heat up
the burner and system. After a period of warm up, the fuel flow
rate is reduced substantially (f/a = 0.03). The continuous
emissions measuring equipment {Bailey and FID) are continuously
monitored to determine when steady state is reached. The
IV-11

-------
thermocouple reading inside the burner is also helpful in this
regard. It generally takes several minutes for most traces of
burner variation to disappear.
Steady state is characterized by a relatively unchanging
reading on the aforementioned instruments (Bailey, FID and burner
thermocouple). The most sensitive instrument for this purpose is
usually the FID. it will detect small changes in burner operation
near blowout well in advance of the other instruments.
When a steady state burning condition has been reached, the
fuel flow rate is decreased in increasingly smaller increments with
allowances made between each change for the burner to come to a
steady state condition. As lean blowout is approached, it will
be noticed that unburned hydrocarbon output begins to increase
as a result of incomplete burning in the burner. However, as long
as the blowout limit is not exceeded the unburned hydrocarbon output
will attain a steady value when steady state conditions exist.
After the unburned hydrocarbon output first indicates that
lean blowout is being approached, the Bailey Heat Prover begins to
show an increase in oxygen content and a measurable reading of
combustibles. This indicates an increase in hydrogen and carbon
monoxide output due to inefficient combustion. The thermocouple for
burner temperature starts to drop.
When the Bailey first begins to show a combustibles
output, a volumetric analysis is taken at each fuel flow setting
(or the gas chromatograph may be used) to determine the composition
of the exhaust stream.
As fuel flow is gradually reduced, a point will be reached
when the burner will eventually go out. This condition is charac-
terized by a gradually accelerating increase in unburned hydro-
carbon output, which never steadies out. When this condition is
noted, a volumetric analysis is rapidly made. The fuel/air ratio
previous to the blowout point is termed incipient blowout and
represents the last steady state burning condition on the rich side
of lean blowout. The lean blowout condition is now straddled by
two points, one on either side of the lean blowout limit.
The air flow rate or inlet temperature is now changed
and the process repeated until the lean blowout limit has been
characterized as a function of the burner inlet condition.
b. Rich Blowout Using Propane
Rich blowout limits are somewhat more
difficult to determine since burning will continue to occur outside
the burner even though the blowout limit has been exceeded within
the burner. The Bailey Heat Power is not used for these tests
since the combustible output is beyond the range of the meter. Some
of the combustibles produced have carbon chains longer than 2 which
gives erroneous results on the Bailey and can coat the detector
with carbon. The FID's use is also limited by the high unburned
hydrocarbon output under these test conditions.
IV-12

-------
Rich operation is characterized by stable but incomplete
combustion within the burner and a large flame outside the burner.
Combustion with insufficient air within the burner produces large
amounts of hydrogen, and carbon monoxide (see Figure 11). There
is usually some hydrocarbon content to the rich exhaust stream
in addition to the H2 and CO.
The products of incomplete combustion burn outside the
burner as they mix with the ambient air, producing a large yellowish
luminous flame.
Detecting rich blowout limits requires some art. It
involves (1) watching combustible temperature in the burner via the
burner thermocouple, and (2) determining whether the burner responds
to reignition at a leaner condition than a suspected rich blowout
condition.
The burner is set up similar to determining lean limits.
When the system is hot the fuel is increased until rich burning
occurs outside the burner as described above. The fuel flow rate is
gradually increased. At each fuel setting, a fuel/air ratio is
determined by taking a volumetric analysis. Instead of increasing
the fuel flow in a straightforward manner, after each fuel setting
the fuel flow is first decreased to the previous setting then
increased to another setting. If the burner has not gone out at a
particular setting decreasing the fuel flow should show an increase
in the burner temperature when the burner is leaned out. If the
burner has gone out at a particular condition, leaning the burner
should show no effect on the indicated burner temperature. Also
if the igniter is initiated at the leaner condition the burner will
reignite (as evidenced by the burner thermocouple output). The
rich blowout limit is thus spanned by a condition where it is
shown that the burner is still lit and a condition where it has
gone out.
Visual observations of burning in the burner and audible
sounds of burner operation are also helpful for absolute assurance
of blowout.
D. Emission Data Collection and Data Reduction
The procedures used to obtain and reduce emission data
will be discussed here. Familiarity with Section V is helpful in
understanding the operation of the equipment discussed. Reference
will be made to Section V when a specific point must be made.
Numerical determination of emission concentration is
either in parts per million (ppm) or grams per kilogram (gm/Kg or
mg/g). PPM data refers to the concentration of pollutants in the
exhaust gas expressed as volume of pollutant per volume of total
flow. This data is most useful in considering emissions from
stationary sources. Gm/Kg data refers to grams of pollutant per
kilogram of fuel. This data is the most significant for mobile
emission sources. Emission data in gm/Kg is necessary to
determine grams of emission per mile.
IV-13

-------
1. Oxides of Nitrogen
a.	Collection Procedure
The oxides of nitrogen concentration
was determined by the Greiss-Saltzman method which is explained in
Section V and Appendix A. Data collection involves evacuating a
flask and then drawing the gas to be analyzed into the evacuated
flask. For the tests, a quartz tube, close coupled to the flask
was inserted directly into the point being sampled; the stack
and/or the burner exhaust. The flask is then sealed.
When a sample is being taken during rich operation, the
evacuated flask is first 1/2 filled with air by connecting it to a
flask of equal size at atmospheric pressure. The sample is then
drawn into the half empty flasli.
To analyze the gas in the flask for NOx, Saltzman reagent
is added, the flask is resealed and shaken for 1 hour to ensure
contact of the gas sample with solution. The solution is then
poured into a cuvette and its optical density at 5500 Angstrom
units is determined, using a colorimeter.
Early in the program, the Saltzman solution was placed in
the flasks before they were evacuated. The sample was then drawn and
the flask resealed and shaken. This procedure produced identical
results to the method previously described.
When a NOx sample after being shaken in the flask for 1
hour produces a very intense color which is known to be saturated,
an additional amount of Saltzman solution is added to the flask and
the shaking period is extended to provide additional dilution and
produce a readable transmission value.
b.	Data Reduction
Calibration of the absorbing solution,
explained in Section V and Appendix A, produces calibration curves
shown in Figures V3 and V4. The best fit through the experimental
calibration data is a family of straight lines on semi-log paper.
The data reduction has usually been done by hand using the curves.
When the computer has been used to reduce the Saltzman data, the
following formula is used.
In(CyRy)
N02 = 5 (1+D) In(R5)
In(I/In)
NOx = 5(1+D) In (I5/I0)
Where:
D = dilution ratio
I = reading obtained after spanning with distilled water
I5 = reading at 5 ppm
IV-14

-------
IQ = reading with pure Saltzman reagent
The data obtained by the data reduction procedures outlined above
is ppm by volume.
2. Hydrocarbons
a.	Collection Procedure
Gas was pumped through a heated pump
and sample line to a flame ionization detector used to detect total
hydrocarbons. The sample gas passes directly through the detector
without going through a packed column, as explained in Section V.
b.	Reference Gases
The instrument was calibrated with
zero air and span gas. The zero is adjusted while zero air is
flowing through the instrument. "Zero air" is not truly free from
hydrocarbons, although every effort is made by the company which
distributes the gas to achieve this result. The degree of impurity
is evident to some extent when the flow rate through the instrument
is changed. Very pure "zero air" shows almost no response on the
FID to a change in flow rate. Zero gases are generally guaranteed
to have less than 0.5 ppm of methane which is equivalent to 0.0 8 ppm
of hexane.
It is conceivable that a really pure gas flowing through
the FID will read less than zero. During the course of the tests
conducted here, zero or slightly negative values were obtained
frequently for hydrocarbon emissions. The zero to the instrument
was then checked for drift by immediately running a "zero air"
sample before the next test point. Whenever drift occurred, an
appropriate correction was made in the data. There were, neverthe-
less, a number of points which gave zeros that could not be
distinguished from the "zero air" setting and this "zero air"
was usually pure.
c.	Sensitivity and Data Reduction
The sensitivity is adjusted while the
span gas is flowing through the instrument. The current practice
is to adjust the instrument flow rates so that a full scale reading
at the most sensitive position corresponds to 10 ppm of hydrocarbon
expressed as hexane. The span gas used normally contains about
200 - 250 ppm hexane equivalent. With this adjustment, the scale
factor for the FID becomes S(FID) = 0.1 ppm/division.
The response of the FID is sensitive to the amount of
oxygen present in the gas stream. This is called the oxygen
synergism effect. With the FID used by Paxve, oxygen decreases
the sensitivity of the instrument to the propane-butane mixture
used for span gas. Figure 12 shows the reduction in sensitivity
of the FID. The attenuation factor is fitted very closely by
Att = 1 - 0.32 tanh
IV-15

-------
where
O2 = oxygen concentration in percent
Final data reducton was accomplished by taking into account the
presence of oxygen in the sampled mixture. This problem is
discussed further in Section V.
3.	Carbon Monoxide and Carbon Dioxide
a.	Collection Procedure
Carbon monoxide and carbon dioxide
were measured by a gas chromatograph using a thermal conductivity
detector, as explained in Section V. Carbon dioxide was measured
by volumetric analysis, also described in Section V. Since the
use of the volumetric analysis is straightforward and was
adequately covered in Section V it will not be discussed here.
Gas samples were pumped from the burner exhaust to a
common manifold and valved from the manifold to each side of the
chromatograph. The sample pump supplies a maximum pressure of 14
psig. During burner operation, sample gas is continuously being
pumped from the burner through the sample manifold, through the
gas sample valve, through the sample loop (see Section V, Fig. 6a)
and is finally vented to atmosphere. During sampling the gas
sample valve is switched so that the sample loop is now ported
on one side to the carrier gas and on the other side to the
column inlet. The carrier gas now flows through the sample loop,
pushing the gas sample into the instrument.
Due to the common manifolding in the instrument to the
column inlets for carrier gas, it is necessary that the sample gas
not enter the instrument at a higher pressure than the carrier gas
pressure. If the pressure in the sample loop is much higher than
ambient, when the gas sample value is switched some of the sample
gas can enter the instrument and travel to the wrong column inlet
(see Figure V-6a). This condition will give erroneous results.
To eliminate this condition the valve on the sample manifold leading
to the sample valve is first closed, so that the gas in the sample
loop returns to atmospheric pressure. Then the sample valve is
switched.
b.	Reference Gases and Data Reduction
Calibrations and zero conditions are
obtained similar to those obtained when detecting hydrocarbons. No
correction must be applied to the data since both sides of the
instrument are linear within the range of interest.
4.	Conversion of Volumetric Data to Gravimetric
Data (ppm to gm/Kg)
It is desirable particularly in the case of
pollutants to express the emission levels in terms of the grams
of pollutant per Kg of fuel. In order to do this we must account
IV-16

-------
for the difference between the molecular weight of the pollutant and
the molecular weight of the exhaust gas. We must also correct for
the water vapor which is formed during the ccpmbustion process but
then condenses in the sampling line before the gas sample is anal-
yzed. An analysis of this problem has been conducted. The result
can be expressed as:
Wp = weight fraction of the pollutant gm/gm
Xp = Mole fraction of the pollutant in the sample
Mp = Molecular weight of the pollutant
K = A constant which depends on fuel and the equivalence
ratio
(a/f)s
4> = Equivalence ratio - (a/f)
(a/f)s = Stoichiometric air/fuel ratio
Xw = Mole fraction of water in the exhaust
Mow = Molecular weight of the wet exhaust gas
Values of Xp are first determined expressed in the ppm. To obtain
the weight fraction of the pollutant in the sample, it is necessary
to multiply Xp by Kp = K Mp to obtain Wp in grams/kilogram of fuel.
Plots of Kp = KMp versus f/a ratio for the various pollutants
produced by propane and kerosene are seen in Figure 13 and 14.
These curves are used to obtain the pollutant emissions in grams/
kilograms fuel.
Wp = KXpMp
Where
IV-17

-------
EMISSIONS SAMPLING
INSTRUMENTATION
CO, 00^,02, HC
OXIDES OF
NITROGEN
ANALYSIS
BURNER TEST STAND 1
CONSOLE
AIR SYSTEM
FIGURE 5
FUEL SYSTEM
FIGURE 5 •
IGNIT,

KM
INSTRUMENTATION
SYSTEM
figures 5;l
*.5
BLOCKHOUSE
BURNER TEST STAND 2
CONSOLE
AIR SYSTEfo
FIGURhi ft
FUEL SYSTEM
FIGURE 6
INSTRUMENTATION SYSTEM
FIGURES 2,4,6
VAPOR GENERATOR SYSTEM
FIGURES 7,fl
IGNITION
SYSTEM
FIGURE
6
FIGURB IV - 1 BURNER EVALUATION FACILITIES

-------
A. Console for Burner and Component Test Facility
(Test Stand Two)
B. Console for Subscale Burner Test Facility
(Test Stand One)
TEST FACILITIES
Figure IV 2

-------
a. Burner/Vapor Generator Installation
In Stand No 2
b. Burner and Vapor Generator System Assembled in Blockhouse
Burner Test Stand No, 2
BURNEfl TEST FACILITIES
Figure IV - 3

-------
B. View of Emission's Sampling and Analysis Equipment
TEST FACILITY FOR GAS EMISSIONS ANALYSIS
A. Overall View of Facility
Figure
IV-H

-------
TEST STAND 1 BURNER SCHEMATIC
FIGURE IV-5

-------
UN5. \\t-KTE_P,
7 V
FUE.L
NOZ7.UL
Gks 6mPLt Pft.o&E.
i,	ST KCV*.
VAPOPi
- Gȣ-NE.PnK"TOR
WOf>NNG F\_U\D
OUT
GK"b SKMPLt.
PP>0&L
BURNER
BURNER TL.ST 5TKND 2.
Schematic Diagram of Fuel & Air Systems
Figure
IV-6

-------
A. Assembled and Insulated Fuel Heater - Vaporizer
B. Disassembled Fuel Heater Vaporizer After Testing
FUEL HEATER - VAPORIZER
BURNER FUEL SYSTEM
Figure IV -7

-------
LIQUID TUEL IK
~
ELECTRICAL FUEL VAPORIZER
FIG. IV 8

-------
a
BURNER TE_VT STKND Z
SCHtfANTlC. D\KGRKM OF VAPOK GtUtRXTO* SYSTE.M

-------

BURNER VAPOR GENERATOR ASSEMBLY
FIGURE IV-10

-------
FUEL AIR RATIO*- f/a
Reference i Combustion of Hydrocarbons— Property Tables	Figure
Purdue University, Eng. Ext. Ser #122, Hay 6fi	IVJn

-------
OXYGEN SYNERGISM EFFECT ON THE FLAME IONIZATION DETECTOR
OXYGEN CONCENTRATION IN THE GAS STREAM 02 PERCENT
0.00	5.00	10.00 15.00	20.00	25.00
OXYGEN CONCENTRATION IN THE GAS STREAM 02 PERCENT

-------
FUEL AIR RATIO, F/a
EMISSIONS NORMALIZING FACTORS FOR PROPANE-AIR
Figure IV-13
0 .01	0.02 0.03 O.OU 0.06 0.08 0.1	0 .2
FUEL AIR RATIO, f/a
EMISSIONS NORMALIZING FACTORS FOR OCTANE-AIR
r l'jre i

-------
TAu^_ ^
BURNER STAND NO. 1
TABULATION OF SUBSYSTEMS AND CONTROLS
SUBSYSTEM
PRIME MOVKR
Air supply system (2) Electrically driven 1.
centrifugal blowezs In 2.
parallel	3.
CONTROL
On-off switch to varlac
Indicator Lamp
Varlac-varlable voltage
to air blower motors
VISUAL MEASUREMENTS
1.	Flowmeter 0-230 pph
2.	Flowmeter Inlet tempera-
ture -75° to +225OF
3.	Air temperature to
burner 0 to 1000°F
Propane supply
system
Kerosene supply
system
Self pressurized propane
tank and electrical
heater
1.	Kerosene tank
2.	Nitrogen supply
pressure cylinder
1.	Propane tank pressure
regulator
2.	Remote solenoid valve on-off
switch
3.	Indicator lamp
4.	Flow control needle valve
1.	N2 pressure regulator
2.	Remote solenoid valve on-off
switch
3.	Indicator lamp
Flow control needle valve
5.	Kerosene tank N2 pressurlza-
tlon valve
6.	Kerosene tank N2 pressure vent
valve
1.	Propane Regulated pressure
0-6 psig
2.	Propane flowmeter 0-8 pph
3.	Propane pressure at Injector
0-30 psig
1.	Kerosene tank regulator
pressure 0-60 pslg.
2.	Kerosene flowmeter
0-8 pph
Fuel thermal
conditioning
1.	Primary heater-
electric Immersion
heating element
2.	Secondary heater exter-
nal resistance heated
tubing
1.	Primary heater
a.	On-off switch
b.	Indicator Lamp
c.	Temperature adjusting
varlac
2.	Secondary heater
a.	On-off switch
b.	Indicator lamp
c.	Temperature adjusting
variao
1.	Temperature at primary
heater outlet 0-1000°F
2.	Temperature at secondary
heater outlet 0-1000°F
""able
IV-i

-------
Table 1 (cont)
BURNER STAND NO. 1
TABULATION OF SUBSYSTEMS AND CONTROLS (CON'T)
SUBSYSTEM
Air thermal
conditioning
PRIME MOVER
Electrical Immersion
heating element
(Paxve) in line
CONTROL
1.	Heater on-off switch
2.	Heater indicator lamp
3.	Temperature control varlac
VISUAL MEASUREMENTS
Air temperature at inlet
to flowmeter -75 to 225°F
Air temperature at inlet to
burner 0 to 1000°F
Burner igniter
High voltage transformer
20 KV/30raa.
1.	Power switch
2.	Indicator lamp
Burner
Fuel/alr/ignlter
subsystems
1.	Power switch
2.	Indicator lamp
Combustion chamber temp.
0-3000°F
Table
IV-l-a

-------
SUBSYSTEM
Air supply
system
TABLE 2
BURNER STAND NO. 2
TABULATION OF SUBSYSTEMS AND CONTROLS
PRIME MOVER
CONTROL-
VISUAL
MEASUREMENTS
2. Electrically driven 1, On-off switch to 1. Flow meter 0-250 PPH
centrifugal blowers
in parallel
varlac
2.	Indicator lamp
3.	Variac-variable
voltage to air
blower motors
Flow meter inlet temp-
erature 0-300° K
Propane
supply
Self pressurized
propane tank and
electrical heater
3.
k.
Propane tank
pressure regu-
lator
Remote solenoid
valve on-off
switch
Indicator lamp
Flow control
needle valve
1.	Propane regulated
pressure 0-100 hSIG
2.	Propane flow meter
0-8 PPH
Kerosene	1. Kerosene tank
supply	2, Nitrogen supply
system	pressurised
cylinder
1.
3.
k.
5.
N^ pressure re-
gulator
Remote solenoid
valve on-off
switch
Indicator lamp
Flow control
needle valve
Kerosene tank
N2 pressuriza^
tion valve
Kerosene tank N2
pressure vent
valve
1.	Kerosene tank regulator
pressure 0-200 PSIG
2.	Kerosene flow meter
0-8 PPH
Air thermal
conditioning
Air heater unit
220 V/AD-20 AMP
Immersion heating
3.
Remote power
contactor
Variac (220 V/AC
2AMP variable
voltage control
to heater
Indicator lamp
1. Temperature at heater
outlet 0-500° F
Fuel thermal
conditioning
1. Fuel vaporizer unit
(Paxve isothermal
heater unit
220 V/AC UO AMP
£ Line heater unit
(Paxve resistance
heating flow element
la Power switch and 1.
remote contactor
b Indicator lamp
c Wheelco pyrome- 2.
trie temperature
controller
2a Multiple coil
current transfor-
mer 7.5 V/AC ?0AMP
b On-off power switch
c Indicator lamp
d Varlac voltage con-
trol to trans-
former 0-12 V/AC
Surface temperature
Paxve vaporizer
heating unit 0-1000°F
Fuel temperature at
burner inlet 0-1000°F
Table Iv_2

-------
Table 2 (cont)
BURNER STAND NO. 2
TABULATION OF SUBSYSTEM. AND CONTROLS (CONTINUED)
SUBSYSTEM
Burner
Igniter
Burner
Vapor
generator
system
PRIME MOVER
CONTROL
High voltage transformer '. Power switch
20 KV/30 ma.
Fuel, air, Igniter
subsystems
1. Working fluid pump
Pump driven Jk HP
VW air cooled
engine
3. Fuel condenser
2, Indicator lamp
Fuel, air, igniter
subsystems
Valve pump bypass
Valve drain or vent
upper
Valve drain lower
Filter total flow
working fluid
2a IGN switch and
lamp
b Cranking switch
and lamp
c Electrical control
to throttle (incr,-
deer.) with lamp
d Pneumatic valve to
engage clutch
e Pneumatic valve to
disengage clutch
VISUAL
MEASUREMENTS
1.	Combustion chamber
temperature 0-3000°F
2,	Exhaust gas (stack)
temperature O-l'tOO0!1'
1.	Pump outlet pressure -
side 0-3000 PSIG
2.	Pump bypass pressure
0-^00 PSIG
(pump side of valve)
3.	Working fluid flow
meter 0-3,5
k. Hydraulic surge
suppressors
a.	Accumulator pump
inlet
b.	Reservoir pump
outlet
5. Vapor generator
heat exchanger
Valve nitrogen supply 1.
to reservoir
Valve reservoir pres- 2,
Bure vent
Valve reservoir isola-
tion
Valve surge chamber
isolation
Valve flow limiting 1.
Condenser inlet pressure
0 - 3000 PSIG
Condenser outlet pressure
-15+60 PSIG
Condenser outlet temp,
0° - 300° F
Reservoir pressure
-15+60 PSIG
Surge chamber pressure
0 - 3000
Vapor generator inlet
pressure
Vapor generator outlet
pressure 0-200 PSIG
Vapor generator inlet
temperature 0-300°F
Vapor generator outlet
temperature 0-1000°F
Table IV-2

-------
V. EMISSIONS MEASURING TECHNIQUES
A. Description and Operation of Instruments Used
During the testing phase of this contract, emission
measurements were made on the burner to determine: (1) the unburned
hydrocarbons, (2) the oxides of nitrogen, (3) the carbon monoxide,
(4) the carbon dioxide, and (5) the oxygen content of the burner
exhaust. The instruments used to determine these emissions are
as follows:
1.	Bailey Heat Prover detected oxygen (O2) and combus-
tibles (a combination of CO and H2).
2.	Volumetric gas absorption analysis detected oxygen
(Oo), carbon dioxide (COo), and on some occasions carbon monoxide
(CO) .
3.	Griess-Saltzman method detected oxides of nitrogen
(NO and NO2) as equivalent nitrogen dioxide (MO2).
4.	Gas Chromatography using a thermal conductivity
detector with thermistor elements detected oxygen (O2)< carbon
dioxide (COj) and carbon monoxide (CO).
5.	A flame ionization detector detected total unburned
hydrocarbons.
The description, operation, and use of these instruments
will be discussed in this section along with the need and use of
auxiliary equipment and calibration techniques.
1. Bailey Heat Prover
The Bailey Heat Prover (Figure 1) is a portable
gas analysis instrument requiring only a source of 10 5-130 volt
ac, 50-60 cycle power. It is otherwise self contained. It is a
continuous reading device which measures oxygen and combustibles.
The instrument is 8.25" deep, 11" wide and 11.25' tall when closed
and 16.5" tall when open and operating. During operation the
instrument must be level, in an area free from drafts and sudden
temperature changes, and provided with an auxiliary filter to
eliminate water and particulate matter which could harm the
instrument.
The instrument has two operating ranges for both the
oxygen and combustibles side: 4% and 20%. When properly
calibrated, the instrument should be accurate to within +0.5% of
the oxygen or combustibles content when on the 20% range and
+0.1% when on the 4% range. These accuracies should not be
expected in the upper or lower 10% of either range.
Gas analysis is performed within the instrument as
follows (see Figure 1-b): A rotary pump having multiDle pumping

-------
chambers draws in air and sample gas at a rate of approximately
100 in3/mm. A large portion of the sample gas is discharged
to atmosphere through a blowoff port. Hydrogen is generated in a
hydrogen cell by the electrolysis of a 10% sodium hydroxide
solution. Hydrogen from the cell passes through the valve block
where it is mixed with a portion of the sample gas and then passes
to the oxygen analysis cell. Another portion of the sample gas is
mixed with air in the pump and then flows to the combustibles
analysis cell.
Analysis in each cell is done having the gas mixture pass
over a noble-metal catalyst which is one leg of a Wheatstone bridge
circuit. The reaction which occurs on the catalyst causes the
temperature to change which in turn unbalances the bridge circuit.
The output of each bridge is proportional to the heat released on
the catalyst and is read out on meters in terms of oxygen and
combustibles.
When properly calibrated, the instrument measures
combustibles directly only when measuring one part hydrogen to
two parts carbon monoxide in nitrogen. Other combustible
mixtures may be determined using the correction curves provided by
Bailey. Large quantities of hydrocarbons with carbon chains longer
than C2 do not give accurate readings and will deposit carbon on
the detector which requires a few minutes of lean operation to
burn off and return the instrument to accurate operation.
During operation, the instrument is plugged into the
appropriate power, the hydrogen generator checked for liquid level
and tightness, the sample jar checked for air tightness, and the
filter checked for cleanliness and dryness.
The instrument is turned on approximately 30 minutes ahead
of analysis by turning the center lever to "check zeros" to allow
for warm-up and to fill the appropriate manifolds with hydrogen.
Before making a test each meter is zeroed and the center lever
turned to "check cell" to determine if the instrument is
reacting properly. At "check cell" the oxygen meter should swing
past the red portion of the meter and then fall back into the red
space. Correct operation is achieved with and the cell adjuster.
After the instrument is operating correctly the center lever is
turned to read the desired analysis position.
During operation, care should be taken to prevent the
meters from remaining in a "pegged" full scale position since this
may overheat and burn out the detectors. Extended continuous
operation should be avoided due to heat build up in some components.
2. Volumetric Gas Analysis
The volumetric gas analysis apparatus shown in
Figure 2 is a cabinet model manufactured by Burrell for flue gas
analysis. The cabinet measures 10.5" wide, 5.75" deep, by 21.25"
high. It is completely portable and self-contained. This model
is set up to detect and absorb, in the following sequence, carbon
dioxide (CO2) in the contact pipette, oxygen (O2) in the first
V-2

-------
bubbler pipette, and carbon monoxide (CO) in the second bubbler
pipette. The liquids contained in each pipette are specifically
compounded to completely absorb only the gas for which it was
designed, i.e. oxygen, carbon dioxide, or carbon monoxide. They all
have low vapor pressures, form stable compounds with the absorbed
gas, and do not outgas other gases.
Operation of this instrument is straight forward. It
requires some manual dexterity and only a little practice to
become proficient with its use. It is necessary to run the analysis
in the proper sequence since the oxygen absorber slowly removes
carbon dioxide and the carbon monoxide absorber slowly absorbs
oxygen. The flushing manifold is provided in the event any of the
solution should enter the manifold. Flushing is achieved by
turning the stopcocks to a through position on the manifold and
flushing it with a %5 sulfuric acid solution. The stopcocks
should be periodically cleaned and greased to prevent gas flowby
through dry or solid build up areas which may develop on the
sealing surfaces.
3. Griess-Saltzman Method
The Griess-Saltzman method is a wet chemistry
technique for detecting the oxides of nitrogen. Paxve followed the
prepartion and analytical procedures, described in Appendix A which
are recommended by the Air Pollution Control District, County of
Los Angeles. The gas sample collection procedure was somewhat
modified as directed by Truesdail Laboratories and will be
described in this section.
The essential equipment necessary for this
method is listed below:
1.	Spectrophotometer (used as a colorimeter)
2.	Vacuum pump
3.	Wrist Action Shaker
4.	Analytical balance
5.	Glassware
a.	Brown ground glass bottles (500 and 250 ml)
b.	1000 ml graduated cylinder
c.	100 ml graduated cylinder
d.	1 liter volumetric flask
e.	50 ml volumetric flask
f.	25 ml volumetric flasks (4 or 5)
g.	10 ml volumetric pipettes
h.	10 ml measuring pipettes
i.	1 ml measuring pipette
]. cuvettes for spectrophotometer
k. collection bottles
(1)
1000
- ml
(2)
300
- ml
(3)
100
- ml
V-3

-------
6. Chemicals
a.	N - (1-naphthyl) ethylenediamine dihydro-
chloride
b.	Sulfanilic Acid
c.	Glacial acetic acid
d.	Sodium nitrite
7.	Tubing for collection bottles
8.	Screw clamps for collection bottles
The sample flasks used are ordinary round bottom flasks
with the neck tapered and attached to a glass tube over which a
tygon tube and screw clamp may be attached. Gas sampling procedure
for NOx varied somewhat during the test program. A discussion of
the problems of NOx measurement is presented in Section V B below.
The final procedure adopted is as follows: A short L shaped length
of quartz tube is inserted into the flask's tygon tube and then
the quartz tube in inserted into an appropriate location of the
burner. The gas sample is obtained by opening the screw clamp and
allowing the gas to enter the evacuated flask. The screw clamp is
then retightened and the quartz tube removed. When the analysis is
to be made, an appropriate amount of Saltzman reagent is added to
the flask and the flask is shaken for an hour to absorb the gas and
develop the color in the dye. After an hour the now colored
absorbing reagent is poured into a cuvette and read on the
spectrophotometer at a wavelength of 550 mp. The spectrophotometer
instrument is first spanned by setting 100% transmission with a
cuvette of distilled water. The oxides of nitrogen is read
directly from either Figure 3 or 4 by entering the appropriate
curve (volume of flask and volume of absorbing reagent used) at
the transmission read. If it is found that the colored absorbing
reagent is too dark to get an accurate reading on the spectro-
photometer (transmission below 10%) and the reagent is not color
saturated, then it may be diluted with distilled water and
then read. The oxides of nitrogen concentration resulting from this
light transmission reading must be multiplied by a suitable factor
to correct for the dilution.
4. Gas Chromatograph Using a Thermal
Conductivity Detector
a. The Instruments
Carbon dioxide and carbon monxide were
separated and detected by using a gas chromatograph using a thermal
conductivity detector. This instrument is manufactured by Carle
Instruments and is designated as Model 8004, Figure 5. It
measures 18.5" deep by 13" wide by 4.75" high with the lid on and 16"
deep by 13" wide by 6.5" high with the lid removed and with a
thermometer installed in the column oven block. Power requirements
are 115 volts ac, 60 H_. For maximum stability the line voltage to
the instrument should be steady, between 100 volts and 125 volts, and
not subject to fluctuation. If the voltage exceeds 125 volts, a
regulating transformer should be used. In use, the instrument
requires, in addition to power, a carrier gas for sweeping the gas
V-4

-------
sample through the columns. Helium is normally used as the carrier
gas due to the large difference between its thermal conductivity
and that of most other gases or liquid vapors. Any cylinder of
convenient size having a regulator capable of delivering gas at
a pressure between 0 and 60 psig is suitable.
The Model 8004 Gas Chromatograph is a dual column, dual
inlet instrument. There are septum covered inlets which allow
samples to be deposited directly at the head of each chromatographic
column. There are also two sample loops which permit a 2 ml gas
sample to be injected into each column by turning a valve. The left
column is a 6 foot long molecular sieve column which is used to
separate O2, N2, and CO. The right column is a 3 foot long silica
gel column which is used to separate carbon dioxide.
Under optimum operating conditions of column temperature
and carrier gas flow rate, this instrument is capable of resolving
5 parts per million of CO2• To obtain the optimum separation, it is
necessary to systematically vary the operating conditions (column
temperature and carrier gas flow rate) and examine the output signal
from the instrument.
The columns are packed 1/8" stainless steel tubes mounted
in contact on either side of an aluminum heater plate which serves
as the column oven and is designed to minimize thermal gradients.
The column oven temperature is adjustable up to 200°C. It is
heated independently of the detector oven. The detector oven is
maintained at a fixed elevated temperature by a constant voltage
supply. The inlet block temperature is controlled by the column
temperature adjustment but is maintained a few degrees above the
column temperature.
The detector is a matched pair of glass coated thermistors
having 100 K impedance that serve as two legs of a Wheatstone bridge
circuit. The thermistors are mounted in lOOy liter chambers to
maintain high resolution and response, in keeping with the small
diameter, low-loaded columns. (Figure 6a).
The output from the instrument is usually fed to a pen
recorder, in this case a Westromcs Model S5E recorder having 6"
wide chart paper traveling at one inch per minute. The response
time is 0.5 seconds full scale. 0.1% full scale resolution is
provided.
Helium carrier gas enters the instrument through a Nupro
metering valve, which is used for fine flow adjustments, and then
flows into the heated inlet block. The column inlets (see Figure
6a) are designed so that the carrier gases flush the annular snace
surrounding the column head eliminating unswept dead spaces. The
helium then flows through the gas sampling valve, and then flows
through the packed column, through the detector, and then
exhausts through 1/8" tubes on either side of the instrument name-
plate .
Two position, six port sample valves are located ahead of
each column. They are designed with minimum valve body volume and
have an optically flat ported stainless steel fixed body that
V-5

-------
s«als against a rotating Teflon coated body. The fixed body con-
nects to the chromatograph and the sample loop by means of 1/16"
stainless steel tubes.
When the valve is in the "load" position (Figure 6c) :
a.	Gas to be analyzed flows into the valve from the
sample pump and is ported through a 2 ml sample loop. The exit flow
from the sample loop returns to another valve port and is vented to
the atmosphere.
b.	Helium flows into the valve from a regulated source and
is ported into the inlet to the column. It passes through the
column and then to the detector. After flowing through the
detector, the helium carrier gas is vented to the atmosphere.
When the sample valve is switched to the "inject" position
(Figure 6d):
a.	The gas from the sample pump flows into the valve and
is immediately vented to the atmosphere. There is thus no
interruption in sample gas flow through the sampling pump.
b.	Helium flows into the valve and is ported into the
sample loop. The return from the sample loop is ported to the inlet
to the column. The sample of gas which was trapped in the sample
loop is thus swept into the column by the helium flow. Flow through
the column is not interrupted, but the 2 ml plug of gas to be
anlayzed is carried in a "sandwich" of helium into the column inlet.
b. Principle of Operation
Separation of gases in a chromatograph
column is a result of the distribution of the gas between two
phases. One of these phases is a stationary bed of large surface
area (the packed column) and the other phase is a gas (carrier gas)
which percolates through the bed. Gas-solid chromatography, the
technique employed here, involves a solid column packing which
separates the gaseous constituents based on their differential
absorption on the column packing. Common packings are silica gel,
molecular sieve, and charcoal.
As gases pass through the column the sample gas mixture
is partitioned between the carrier gas and the solid stationary
phase. The solid phase selectivity retards the sample components
according to their distribution coefficient until they form separate
bands in the carrier gas. These component bands leave the column
in the gas stream and produce an output signal from the detector
based on each components thermal conductivity. The fluctuation in
the output signal produces a series of peaks on the recorder paper
according to component bands of gases leaving the column.
The record of the output signal is termed a chromatogram,
and the time from the start of the analysis to the time any gas
component is eluted is called the retention time. The area under
each peak on the chromatogram is proportional to the concentration
V-6

-------
of the corresponding gas in the sample mixture. The area may be
determined by any number of methods. The peak height gives a good
indication of the area when the peak is symmetrical, tall, and thin
and produces good results when the instrument has been calibrated
in terms of peak height. The majority of chromatographers use peak
height methods in preference to any other integration technique
(Reference 1) even though it is not the most basic method.
To maximize the peak height for a particular gaseous
component, a gas mixture having the component in the concentration
desired is run through the instrument while the column temperature
and carrier gas flow rate is systematically varied. The carrier gas
flow rate is accurately determined by attaching a soap bubble flow
meter to the tube leading from the detector. The meter consists of
a vertical tube with etched marks 10 ml apart and a reservoir of
soap at the bottom. The carrier gas is bubbled through the soap and
the flow rate of the gas is measured by timing the passage of the
bubble between the two marks. Figure 7 shows the results of such
an optimization procedure. From this figure the instrument was
operated with a column temperature of 100°C and a carrier gas flow
rate of 15 cc/minute.
c. Operating Procedure
Emissions measurements procedures
with the thermal conductivity gas chromatograph for carbon dioxide
and for carbon monoxide are substantially the same. To put the
instrument into operation, the Helium carrier gas flow is first
started to prevent oxidization of the column material. The
electrical power to the instrument is then turned on and the column
oven temperature control adjusted to obtain 100°C. The instrument
is allowed to warm up for at least one half hour before further
adjustments are made.
The instrument is operated at 100°C because, when combined
with an appropriate carrier gas flow this produces an optimum
carbon monoxide separation and peak height. A thermocouple is
fitted into the column head to monitor its temperature.
After the instrument is up to temperature, the
chromatograph attenuation switch is put on "test" and the recorder is
zeroed. The attenuation switch is then put on "1" and the recorder
is zeroed by adjusting the coarse and fine zero adjustment
potentiometers on the instrument body. The instrument is now ready
for use.
During use, samples are pumped to a common manifold and
are valved from the manifold to each side of the chromatograph. The
sample pump supplies a maximum pressure of 14 psig. During burner
operation, sample gas is continuously being pumped from the burner
through the sample manifold, through the gas sample valves, through
the chromatograph sample loops. In order to take a reading, the
sample flow from the manifold is shut off, and the gas in the
sample loop of the chromatograph is allowed to blow down to
atmospheric pressure.
V-7

-------
The gas sample valve then is switched so that sample loop
is ported on one side to the carrier gas and on the other side to
the column inlets. The carrier gas now flows through the sample
loop, pushing the gas sample into the instrument.
The blow down of pressure in the sample loop is required
in part by the design of the dual column chromatograph. Due to the
common manifolding in the instrument column inlets for carrier gas,
it is necessary that the sample gas not enter the instrument at a
higher pressure than the carrier gas pressure. If the pressure in
the sample loop is higher than the carrier gas pressure when the gas
sample valve is switched, some of the sample gas will blow back
through the carrier gas line to the wrong column inlet (see Figure
V6a). This condition gives erroneous results. To eliminate this
condition the valve on the manifold leading to the sample valve is
first closed, so that the gas in the sample loop returns to
atmospheric pressure. Then the sample valve is switched.
5. Hydrocarbon Analysis Using a Flame
Ionization Detector
a. The Instrument
Total hydrocarbon emissions were
determined using a flame ionization detector. The instrument used
for this purpose was a gas chromatograph manufactured by Carle
Instruments designated the Model 9000. The instrument measures
9" high by 19.875" deep by 16" wide with the lid on, and 6.5"
high by 18.75" deep by 16" wide with the lid removed and the
instrument in use. Power requirements are similar to those of
the thermal conductivity gas chromatograph. In addition, however,
two polarizing batteries, each 300 volts must be used. These need
to be replaced approximately once a year.
This chromatograph is normally used as a dual flame,
dual column FID gas chromatograph instrument with the dual-inlets
extending into the columns. The columns are heated by a central,
alumimun plate, sandwich-type heater designed to minimize
gradients in the column. The column temperature can be varied
between ambient and 200"C. The detector oven is heated separately
by a small cartridge heater maintained a few degrees above the
column temperature by a common variable transformer setting.
The instrument used by Paxve used only one column. It is
set up to measure total hydrocarbons in the left hand detector and
to separate and identify hydrocarbons in the right hand detector.
Gas samples entering the total hydrocarbon side of the instrument
pass continuously through a flow restrictor (needed to balance
the restriction due to the column on the other side of the
instrument) and then to the detector. When using the instrument as
a total hydrocarbon analyzer, emission data is obtained continu-
ously, as opposed to the sampling techniques needed for CO, CC>2 and
hydrocarbon separation.
Gas samples entering the separation side of the instrument
pass through a gas sample valve where the gas is continuously
V-8

-------
flushed through a sample loop. For a gas analysis, the valve is
switched so that carrier gas is flushed through the loop and the
sample gas is sent through the column for separation and finally
to the detector for readout. The column used in this instrument is
a 24 foot, 1/8" OD stainless steel tube packed with DC 200, a
material chosen for separating saturated hydrocarbons (although
equally efficient for aromatic and olefinic hydrocarbons).
The Model 9000 FID chromatograph requires three gases
for operation; a carrier gas for flushing the sample gas through
the column, and hydrogen and air for the flame ionization detector.
When the instrument is used as a total hydrocarbon detector, the
carrier gas is only needed to prevent air from entering the
separation column and oxidizing its surface.
b.	Principle of Operation
The detector consists of dual hydrogen
flames enclosed in a teflon and metal walled chamber (Figure 6b).
Sample gases continuously pass through the hydrogen flame, and
produce ions. The electrical conductivity of a hydrocarbon free
hydrogen flame is very small. When hydrocarbons in the sample gas
pass through the flame the conductivity of the flame rises due to
the creation of charged particles (electrons and negative and
positive ions) with the result that a current will flow from the
charged plates to the input electrodes. To achieve a low background
level, it is necessary to supply the instrument with so called 'zero
gases" which are virtually free of hydrocarbons.
The current which flows across the charged plates produces
a voltage drop across a resistor in an electrometer (a portion of the
electronic circuitry) which is amplified and then fed into a
recorder. The detector operates in the same way regardless of
whether the instrument is being used to record continuous data or
sampling data.
c.	Operating Procedure
During hydrocarbon emissions testing
the instrument was operated at 200°C in keeping with the heated
sampling line leading to the instrument. A thermometer fitted into
the column head is used to determine the instrument temperature.
To operate the total hydrocarbon side of the instrument,
carrier gas flow is started to prevent oxidization of the column.
The instrument is turned on and the temperature adjusted until the
thermometer reads 200°C. Air flow is turned on and adjusted to
about 800 ml/min (excess air flow). The hydrogen is now turned on
and adjusted to about 20 ml/min with the bubbler flow meter. The
hydrogen is now lit and the detector housing returned in place and
allowed to come to temperature equilibrium.
Optimum hydrogen flow is related to flow of gas to be
analyzed. Optimum conditions exist when hydrogen flow rate = 1.1 x
sample flow rate. Since the flow restrictors are located in the
oven, all gaseous flow rates are temperature dependent. Figure 8
V-9

-------
shows the calibration curves which were determined for this
instrument and were used to determine the appropriate flow rates.
The instrument is adjusted by first balancing the
amplifier for zero output. The instrument is then zeroed by running
"zero air" through the instrument at the same rate as the sampling
rate. The instrument is then spanned by passing a hydrocarbon
mixture of known concentration through the instrument. The sample
flow rate was set at this point by adjusting the flow rate of span
gas mixture (containing 2 39 ppm of HC expressed as hexane), so
that the output of the recorder is 2390 scale divisions. The
recorder now will register 10 scale divisions for every ppm of
hexane in the sample gas. The sampling pump bypass is then adjusted
so that it delivers the same flow rate as the span gas.
B. Instrument Calibration
1.	Bailey Heat Prover
The Bailey Heat Prover reads both oxygen and
combustibles. The oxygen calibration is built into the instrument.
Adjustments for both the 4% and 20% full scale settings are provided
together with a built in zero adjustment.
The Bailey Heat Prover combustibles meter is factory
calibrated. A calibration may be made using a Heat Prover Checker.
In essence, the calibration consists of flowing a gas of known
composition through the instrument and adjusting calibration
resistors so that the instrument reads the given gas composition.
The calibration technique results in producing a one to one
relationship between the indicated combustibles and the actual
combustibles when the ratio of carbon monoxide to hydrogen is 2
to 1. Other carbon monoxide to hydrogen ratios and other combus-
tibles will not produce a one to one ratio of indicated to actual
combustibles. Curves giving correction for other combustibles
are given by Bailey for various gases and mixtures. These curves
were used when the theoretical total indicated combustibles data
shown in Figure IV-11 was plotted.
Periodic factory calibrations for combustibles were
performed during the testing period. Frequent checks for proper
oxygen calibration were made during each day.
Daily minor adjustments as indicated in Secion V-A-l are
necessary to keep the instrument in factory calibration. Other
adjustments may be needed as the instrument is used. The procedure
for the many other adjustments may be found in the operation manual
E65-15.
2.	Volumetric Gas Analysis
The volumetric gas analysis equipment requires
no calibration. Periodic checks are sometimes run when it is
suspected that one of the fluids has reached its saturation. To
obtain complete absorption of a particular gas component, it
V-10

-------
is usually necessary to make two or three passes through a gas
absorption pipette. if more than three passes are required for
complete absorption, the reagent is becoming saturated. Once a
particular reagent becomes saturated, it is necessary to replace it.
3.	Oxides of Nitrogen
The absorbing reagent is prepared per the
procedure outlined in Appendix A. According to Saltzman (Appendix
A) and the ASTM procedure D 1607 "Nitric Dioxide of the Atmosphere",
the absorbing reagent is stable for several months if kept refriger-
ated and well stoppered in brown bottles. The absorbing reagent is
calibrated with a standard sodium nitrite solution.
Sodium nitrite solution is added to the absorbing
reagent so as to make a total mixture of 25 ml. This is then shaken
and allowed to stand for 15 minutes. The NO2 in the NaN02 reacts
with the chemicals in the solution to produce a dye which is
reddish purple in color. The dye has a strong light absorption
peak at 550 my. Increasing the amount of sodium nitrite added
caused increasing dye intensity until a maximum is reached when
all of the dye has been used up. This condition is known as
'saturation' of the Saltzman absorbing reagent. Further addition
of NaN02 causes the light absorption to decrease due to dilution.
A curve of light transmission versus equivalent
concentration of nitrogen dioxide in micrograms per 10 ml of
absorbing reagent is obtained from the NaN02 data. This curve
is then used to draw curves for determining the concentration of
NO2 in the gas samples. It is also used to determine the point at
which the absorbing reagent becomes saturated with regard to
the maximum amount of nitrogen dioxide it can detect.
The NaN02 light transmission curve and formula (1) from
Appendix A, are used to draw families of curves for various flask
sizes and reagent quantities from which the nitrogen oxides
concentration in ppm may be read directly as a function of light
transmission (Figures 3 and 4). Flask sizes were chosen so that
it was possible to detect nitrogen oxides from 0.1 ppm up to 200
ppm, with a reasonable range of sample flask size and volume of
absorbing reagent required.
4.	Gas Chromatoqraph Using a Thermal
Conductivity Detector
This instrument is used to separate and measure
carbon dioxide (on the right side of the instrument) and carbon
monoxide (on the left side of the instrument). Since both sides
of the instrument are similar in operation, only the calibration
of the carbon monoxide side of the instrument will be described
in detail.
a. Zero and Span Gas Calibration
After the instrument is set up
V-ll

-------
according to the directions in Section V-A-4, it is zeroed by
setting the attenuation to 1 and zeroing the recorder pen with the
zero controls. A span value is now obtained by introducing into
the instrument through the sample loop, a span gas mixture having
a known concentration of carbon monoxide.
The span gas is a mixture of gases designed to permit
calibration of all of the chromatographic equipment with the same
gas. The span gas mixture used for this purpose by Paxve has the
following nominal compostion
200 ppm-propane
200 ppm-n-butane
300 ppm-carbon monoxide
5% -carbon dioxide
balance-nitrogen
The span gas produces a scale reading on the chroma-
tograph recorder when the gas in question is eluted from the column.
Dividing the known carbon monoxide concentration by the recorder
deflection produces a scale factor. This scale factor is then used
to multiply all subsequent scale deflections to obtain carbon
monoxide concentration of any unknown sample gas being analyzed.
The scale factor may change due to internal changes in the
chromatograph during testing. For this reason, span calibrations
are made frequently during the course of testing and the scale
factor noted.
b. Linearity of the Instrument
The use of a scale factor as described
above assumes that the instrument is linear over the operating range
of interest. This question, the linearity of the thermal conduct-
ivity gas chromatographs, was investigated by means of a dilution
test procedure which is described in Section 6 below. That
test procedure allowed us to put known concentrations of gases
to be anlayzed through the chromoatograph and to observe the
resulting output signal as read on the recorder. We were able to
explore the entire range from a pure gas down to the lowest
concentration that could be distinguished on the chromatogram.
Figures 9 thru 11 show the results of the dilution tests
for carbon monoxide, carbon dioxide, and oxygen. It was found that
both the carbon monoxide and carbon dioxide peak heights were in
fact linear with concentration over most of the range, to within the
accuracy of the experiment. Oxygen concentration was not linear
with peak height, showing an increasing sensitivity as the concen-
tration was reduced. Since the chromatograph was not used to
analyze for oxygen, this nonlinearity was not significant for the
work reported here.
5. Hydrocarbon Analyzer Using Flame Ionization
Detector
The Carle Model 9000 gas chromatograph is used to
detect and measure total hydrocarbons by passing the sample gas
V-12

-------
directly into the detector, without going through a packed column,
as explained in Section V-A. Calibration of this instrument is
virtually the same as calibrating the thermal conductivity gas
chromatograph with the following exceptions:
(1)	The FID gas chromatograph is a continuous
reading instrument so that gas samples are
not injected directly into a column with
a valve, and
(2)	The response of the instrument's output to
a particular gas concentration can be
varied by adjusting the delivery pressure
and hence the flow rate of the unknown
gas into the detector.
a.	Zero and Span Gas Calibration
After the instrument is set up, zero
air is flowed through the detector with the attenuator settings both
on 1. The recorder pen is zeroed by adjusting the coarse and fine
suppression controls. It may be necessary to adjust the polarity of
the suppression controls to obtain a zero. The instrument was then
spanned by running a span gas mixture through the instrument and
adjusting the delivery pressure so that 1 scale division was
equivalent to 0.1 ppm hydrocarbon expressed as hexane. The zero and
span are now re-adjusted if necessary to obtain consistent values.
"Zero air" is not truly free from hydrocarbons, although
every effort is made by the company which distributes the gas to
achieve this result. The degree of impurity is evident to some
extent when the flow rate of zero air through the instrument is
changed. Very pure "zero air" shows almost no response on the FID
to a change in flow rate. The gas is generally guaranteed to have
less than 0.5 ppm of methane which is equivalent to 0.08 ppm of
hexane. It is conceivable therefore that a really pure gas flowing
through the FID will read slightly less than zero.
During the course of the tests conducted by Paxve, zero
or slightly negative values were obtained frequently for hydrocarbon
emissions. The zero of the instrument was then checked for drift
by immediately running a "zero air" sample before the next test
point. If drift had occurred, an appropriate correction was made in
the data. There were, nevertheless, a number of points which gave
zeros that could not be distinguished from the "zero air"
setting. There were also many readings which were definitely
negative.
b.	Linearity of the Instrument
Dilution calibrations described m
Section 6 were run for the FID gas chromatograph. These were run for
both the left hand and the right hand sides of the Model 9000.
Dilution by both air and nitrogen were used.
It was found from the data for the right hand side of the
Model 9000 (which separates the gases with a column) that the
V-13

-------
detector itself is linear with concentration for both propane and
butane over the entire range of the instrument. This is shown in
Figure 12.
Flow through the left hand side of the Model 9000 FID
showed that the detector was sensitive to the presence of oxygen.
When N2 was used to dilute the span gas, the response of the totals
side of the instrument was linear with the span gas content. When
zero air was used as the diluent, the response of the instrument was
not linear. There is a reduction in sensitivity of the instrument
with increasing oxygen concentration.
The apparent nonlinearity of the FID with O2 present is
in fact not a nonlinear response to hydrocarbons. Rather, it
is a suppression of FID sensitivity to hydrocarbons in the
presence of oxygen. This suppression is called oxygen synergism.
It is discussed in more detail in Secion V-C-2-c of this report.
Of particular interest here is the magnitude of the
suppression as a function of oxygen content. Figure IV-12 shows
the ratio of the FID response with air dilution versus its
response with N2 dilution, plotted against C>2 concentration
We see that for O2 of approximately 10%, the ratio is 0.76.
c. Zero Shift
The negative readings of the hydro-
carbons emissions from the burner was a continual problem area during
burner emissions testing. After the fuel injector problems on Stand
2 were cleared up (Runs 282 and later), almost all normal runs
showed hydrocarbon emissions which were in the range from +0.5 ppm
to -0.5 ppm HC (5 scale divisions on the recorder) expressed as
hexane.
The possibility that water vapor was the cause of this
problem was investigated. It was found that water vapor at about
10% concentration in air did not influence the span gas reading of
the instrument. It was also found that 10% water vapor did produce
about a -5 scale division shift in the recorder output, equivalent
to -0.5 ppm of hexane.
Paxve has not made a systematic investigation of the
influence of water vapor on the FID. We believe that the zero shift
which we observed is the probable explanation for the negative
readings which were obtained during burner operation. We did not
devise or employ any calibration technique in this regard, but we
feel that this is an important area for further work. This becomes
increasingly desirable with the development of unusually low
emission burners such as the Paxve burner being reported on here.
6. Dilution Calibrations
In using a scale factor we have assumed that the
response of the chromatograph is linear with concentration. Thus
we assume that a CO concentration of 300 ppm will give twice the
recorder deflection as a CO concentration of 150 ppm. In order
V-14

-------
to investigate the linearity of the instrument, a range of
known mixtures must be analyzed. Two dilution techniques were used
to achieve this information:
(1)	a dynamic dilution technique using an
"exponential dilution flask".
(2)	A static dilution technique where a gas
mixture of known concentration was
repeatedly diluted in a determinable
fashion.
In each case the resulting mixtures were then fed to the instrument
whose linearity was being examined.
a. Dynamic Dilution
The dynamic dilution tests were
conducted using an "exponential dilution flask" of known volume
in which a gas of known composition was diluted continuosly by a
measured flow of a diluting gas. The exponentially varying
concentration of the resulting mixture was then analyzed measured by
the instrument at measured time intervals.
The change in concentration of the calibrating gas being
drawn from the dilution flask with time is given by
-tw
C = C0 e ~
where
t = time from start of run - sec
C = concentration at time t - ppm
CQ = concentration at time t = 0 - ppm
v = volume of the mixing flask - cc
w = flow rate of diluent - cc
sec.
A schematic diagram of the exponential dilution test
apparatus is shown in Figure 13. As shown in this schematic diagram
the typical calibrating gas storage bottles consisting of span gas,
zero air, pure nitrogen, carbon monoxide, and carbon dioxide are
connected by appropriate valving to a supply manifold which can be
individually valved to the exponential dilution flask. A given
calibrating gas can be locked into the flask through an appropriate
set of valves. The diluting gas flow may then be set and the
apparatus connected to the particular chromatrograph channel under
evaluation.
At the start of the test, the diluting gas is switched
into the flask at a known flow rate and an exponential dilution of
the calibrating gas in the flask then takes place which is analyzed
at timed intervals by the chromatograph. In this manner a
continuously decreasing concentration of the original gas in
the flask can be evaluated over the full range of the instrument.
It was found that heat must be added to the flask as the dilution
process proceeded in order to maintain a constant stirring of the
V-15

-------
gas molecules and to prevent surface adsorption of the gas molecules
on flask walls.
Results of the dilution sensitivity test of the Carle 9000
FID gas chromatograph are shown in Figure 14. This test was
conducted with span gas diluted with zero air using a heated flask.
The effect of an unheated flask is shown in Figure 15. It is
evident from Figure 15 that deviations in linearity were appreciable
at both the extreme limits of the span when no heating was provided.
This was not the case, as shown in Figure 14, when the flask was
heated.
b. Static Dilution
The static dilution technique involved
filling a sample cylinder with span gas or with pure gas (carbon
monoxide or carbon dioxide), depending on which instrument is to
be calibrated, repeatedly diluting the sample, and then analyzing
the cylinder contents. The procedure was initially started with the
flame ionization chromatograph. Flasks were filled with samples of
span gas and accurately diluted with nitrogen or zero air to obtain
several different known span gas mixtures. These mixtures were then
run through both the totals (left side) side and the separation
(right side) of the Model 9000 FID chromatograph.
The resulting curves Figures 16 and 17, show that when
nitrogen is used as a diluent, both sides of the instrument are
linear with concentration. When air is used as a dilutent, the right
hand side of the instrument (separations side) remains linear
(Figure 12), while the left side of the instrument shows a
nonlinearity. Since the right side sees the constituents
separately, it is clear that the nonlinearity on the left hand side
is due to the simultaneous presence of O2 and hydrocarbons. This
is the oxygen synergism effect discussed in Section V-C-2.
A similar dilution procedure was followed to obtain
gas mixtures for the thermal conductivity gas chromatograph. The
proven linearity of the FID right side was used to assist in the
analysis. Gas compositions having concentrations less than 300
ppm for carbon monoxide and 5% for carbon dioxide were obtained by
diluting span gas with zero air or nitrogen. Gas compositions
having higher concentrations of CO and CO2 were obtained by
diluting pure carbon monoxide or carbon dioxide with span gas. In
either case, the concentration of the diluted gas mixture was
determined from the concentration of the hydrocarbon in the
mixture. This was in turn obtained from the FID.
An oxygen calibration for the left side of the thermal
conductivity gas chromatograph was accomplished using the static
dilution technique. Zero air diluted with span gas was found to
be unsatisfactory for this purpose because the high concentrations
of nitrogen caused the leading edge of the nitrogen peak to overlap
the oxygen peak in the chromatogram. This in turn gave rise
to an apparent nonlinearity in the oxygen calibration. To
avoid this problem, a 50/50 mixture of air and span gas was diluted
with carbon dioxide. The hydrocarbons in the span gas were used to
V-16

-------
establish the composition of the mixture. This technique yielded
satisfactory separation of the oxygen and nitrogen peaks.
Figures 9, 10 and 11, show the results of the Model
800 4 gas chromatograph linearity investigations. The CO and CO2
responses are seen to be linear. The oxygen response shows
nonlinearity of peak height at high concentrations.
C. Emission Measuring Problems
During the course of this program many unforeseen problems
arose in connection with the emission measuring equipment and the
sampling techniques used in conjunction with that equipment. Some
of these problem areas are treated in the technical literature, but
they were unknown to us at the beginning of the program and were not
brought to our attention by the manufacturers of the emission
equipment we purchased. Other problem areas are either unknown to
workers in this field or are merely considered part of the "art".
They are not discussed in any available literature. These problems
and our solutions to them are reported here in the hope that others
may profit from the work.
In addition, data reduction techniques not discussed
elsewhere have been included here as an aid in understanding the
reduced data presented in Section VI.
1. Oxides of Nitrogen
The oxides of nitrogen are determined by a wet
chemistry method known as the Griess-Saltzman method. This method
is described in Section V, A3, B3 and Appendix A. During the course
of the program the testing procedures were modified as new knowledge
was gained.
Initial NOx data was gathered by sampling from a manifold
(Figure IV-4 B) used to feed the TC gas chromatograph. This manifold
was the terminal end of a 52 foot sample line carrying gases from
the burner on stand 1 to the chromatograph (see Figure 19). The
initial NOx results using gas from this sampling point collected
in 1000 ml sampling flasks with 10 ml of absorbing reagent are
shown in Figure 18 labeled "Initial Data". The highest NOx values
obtained during these tests were 17 ppm.
Suspicion was raised about the flat top and overall shape
of the curve of NOx compared to the equilibrium NO values shown. As
a result, some studies were initiated which revealed three problem
areas in NOx measurement. These problems are indicated by a, b,
and c in Figure 18. They are discussed below.
a. Absorption of NOx by Condensed
Water
As explained earlier, NOx data
gathered from a remote sampling point was considered suspicious. It
was initially thought that the hot tip of the metal sample line
might be catalyzing a reaction which the NO2 or NO was breaking
V-17

-------
down to form compounds which were undetectable by the absorbing
reagent. A test was run where gas samples for NOx analysis were
taken from several points (labeled #1, 2, 3, and 4 in Figure 19)
using 2 sample probe materials.
Gas samples were taken from the burner directly (point 1)
using a 1 foot length of stainless steel line for a sample probe and
a 1 foot quartz tube sample probe. For the same set of burner
operating conditions, samples were taken from the sample line at
points, 02, #3, and #4. This last location was the sample point
at the manifold which was used to gather the HOx data in the
earlier test runs. Table 1 shows the results of these tests.
It is seen that there is virtually no difference in the
NOx for the two sampling materials when sampled at point #1. The
idea that stainless steel destroys or otherwise alters the MO or
NC>2 to produce low readings is clearly not supported by this data.
As the sampling point was moved away from the burner
the measured NOx concentration decreased. This reduction in
detectable NO2 can be attributed to the absorption of MO2 in the
condensed water vapor which drops out in the sample line.
Nitrogen dioxide (N02) dissociates in water according to
2	NO2 + H20 -~ HHO3 + HN02 (cold)
3	NO2 + H20 -»¦ 2HHO3 + NO (warm)
Nitric acid is infinitely soluble in water and therefore remains in
solution. Nitric oxide (NO) is only slightly soluble in water
(.0059 gms NO dissolve in 1 gm water at 23°C and 760 mm Hg) so
that some of it will become dissolved in the water and some of it
will come out of solution and enter the flowing gas stream. In
the gas stream nitric oxide (NO) is slowly oxidized by the available
oxygen to nitrogen dioxide which is further trapped in the water as
explained above. Except for being slowly oxidized to nitrogen
dioxide, as explained above, nitric oxide is recovered in the
sample gas and detected by the Griess-Saltzman method.
The loss of H02 as opposed to NO in the long sample line
could be readily observed when the sample flasks were developed on
the shaker. When samples were taken from the long stainless steel
line a considerable period of time (about 15-30 minutes) was
required before any color developed in the absorbing reagent.
This time period was a result of the NO being slowly oxidized to
NO2. When samples were taken with the short quartz tube, color
began to develop almost immediately due to the presence of NO2-
As further substantiation of the absorption of NOx by
condensed water, a sample of the condensed water was analyzed by the
Griess-Saltzman method and was found to contain 18y grams of
equivalent NO2/10 ml of absorbing reagent. Although nitric
acid is not supposed to produce a positive response with the
absorbing reagent (See Saltzman, Appendix A), nitric acid is
unstable and breaks down under sunlight and also heat according to:
V-18

-------
4 HNC>3 -»¦ 2 H2O + 4 NO2 + O2 +
Nitric acid is a powerful oxidizing agent and decomposes in the
presence of a reducing material according to:
2 HNO3 ->• H2O + 2 N0+ + |£>2 +
In any event a compound which is detected by the absorbing reagent
(NO2) or readily converted to such a compound (NO) is formed.
•
A simple analysis was also performed to show the
equilibrium concentrations of the various species of nitrogen
oxides with water.
let a = solubility of NO in water gm/gm Atm
[NO]i = concentration of NO in water gm/gm
[NO]g = concentration of NO in gas gm/gm
pNOG = Partial pressure of NO m gas Atm
MWW = Molecular weight of water
MWN0 = Molecular weight of NO
mwEX = T^e molecular weight of exhaust gas
Wi = Weight of NO in liguid per pound of exhaust
Wg = Weight of NO in gas per pound of exhaust gas
XW = The mole fraction of the exhaust gas which is water
XNO = The mole fraction of the exhaust gas which is NO
Then if most of the NO is in the gas
PN0 = XN0
[NO] = PNO a
MWNO
[N0]g = PNO MWex
The weight of No in the water per pound of exhaust will be
MWW
WL = [NO]x X W MWex
The weight of NO in the gaseous phase per pound of exhaust will be
W„ = [NO]„
V-19

-------
The fraction of the NO in the liquid phase is then
Wt g Xw MW
Wg = MWNO
Substituting
a = 5.6 x 10~5 gm/gm
Xw = 0.10
MWW = 18
MWNO =28
we obtain
= 3.6 x lO"6
thus the amount of NO which is lost to the water which condenses
in the line is negligible.
Our interpretation of the experiments and the analysis was
that we were losing the NO2 in the line, but not the NO. This of
course did not consider the possiblity that some of the NO was
converting to NO2 in the line. The residence time in the line,
however, did not seem sufficient for that reaction to proceed.
As a result of the tests described above, the sampling
procedure for oxides of nitrogen was modified. All test data
for oxides of nitrogen taken after run 10 3 used a short quartz tube
which drew the sample to be analyzed directly into an evacuated
flask. Some of the early data gathered using this procedure was
plotted in Figure 18. The comparsion between the new data and
the old data is clear. The new data still shows the flat top exhib-
ited by the old data (peak values of NO always less than or equal to
17 ppm). It did, however, show larger values of NOx at low fuel air
values.
Superimposed on Fig. 18 is a curve of the theoretical
NO2 which exists in equilibrium in the exhaust as a function
of the fuel air ratio. We see that the difference between the
new NOx values and the old NOx values is approximately the
same as the theoretical NO2. From this we might infer that the
NO2 in the burner is approximately in equilibrium, and that
when we lost it in the lines, we were losing NOx of at most
3 ppm. For most burners, a combustion exhaust gas analyses
whicl; failed to find a 3 ppm of NOx would not be considered a
serious problem. In the case of the Paxve burner at low fuel
air ratios, however, this may represent 50% or more of the
entire NOx in the exhaust.
b. Saturation of the Absorbing
Reagent
It was apparent from the flat top of
V-20

-------
the NOx data curve that some type of saturation phenomenon might
be occurring. Another possibility that was considered was that the
Saltzman solution responds differently to NO or NO2 gas than it
does to the NaN(>2 calibrating solution. This latter idea was not
considered too seriously, since it was precisely this problem
which Saltzman investigated. Nevertheless, we decided to attempt
a direct NO or NO2 calibration of the Saltzman reagent, in spite
of the considerable difficulties involved.
Mixtures of NO (which contained some NO2) in dry Nj were
prepared which should have been about 100 ppm. The Saltzman reagent
only indicated 23 ppm. When a similar mixture was prepared by
collecting the gases over water, the Saltzman reagent showed almost
no color.
Finally mixtures of NO2 in air were prepared. A
mixture which contained approximately 5 ppm indicated 5 ppm in the
Saltzman reagent, but a 10-0 ppm mixture only indicated 19 ppm.
It was clear that saturation of the reagent was occurring.
It was also clear that the reagent was sensitive to NO2, but did not
respond to NO. The saturation phenomena was further investigated by
means of the NaN02 solution.
The standard procedure for calibrating the Saltzman
solution involves making up a mixture of Saltzman reagent and
NaN02 standard solution of total volume 25 ml. We normally use
from 0.2 ml to 6 ml of the NaN02 with the balance made up of
Saltzman reagent. The optical transmission of the resulting
solution plots as a straight line on semi-log graph paper,
indicating that the transmissibility is given by
ln = K1 p
= K2 [no2]
I = light transmission at 550 m
Io = transmission through clear reagent
a = the concentration of the dye in the liquid mixture
[NO2] = the concentration of N02 in the mixture
K1 k2 = appropriate constants
As long as there is enough dye to indicate all of the N02
that is present, we should expect
Where:
ln (fe) = k3[no2
„ VN K±
"K4VN+VS = 25 VN
V-21

-------
where
VN = volume of NaN02 solution
Vg = volume of Saltzman reagent
If however, we continue to increase the proportion of
NaNC>2 solution until there is an excess of NO2 and insufficient dye
available, we would then expect the transmission to be given by:
I_
In I0 = k1Pd
VS
= K4 Vs+Vfj
Figure 20 shows the light transmission data obtained by
varying the amount of NaN02 solution from .2 ml to 20 ml. The data
is well fitted by two straight lines on semi-log paper. The left
hand line represents the usual calibration curve for the Saltzman
solution. In this region, there is an excess of dye and all of the
NO2 present is indicated. The right hand line represents the
saturation region. In this region, there is a deficiency of dye
and only the amount of dye present is indicated. This behavior
is exactly that predicted by the above analysis.
By finding the slope of the saturation line in the graph,
the maximum dye concentration at saturation can be found. From
this we can deduce what concentration of NO2 will yield saturation.
In the case shown here, the dye saturates if it is exposed to 21yg
per 10 ml of NO2. This is equivalent to approximately 19
ppm of NO2 in a 1 liter flask containing 10 ml of reagent.
Much of our early data showed values which were
between 12 ppm and 17 ppm at conditions where higher values
were expected. We therefore decided to disregard any early
data over 10 ppm as possibly saturated. We also decided to
change our procedure to avoid saturation.
From Figure 20 it was found that the absorbing reagent
saturates at approximately 21p gms of N02/10 ml of absorbing
reagent. If this number is substituted into formula (1) of the
procedure for preparing the absorbing reagent (Oxides of Nitrogen,
Appendix A) the following results:
(0 . 532 ) (21 qui/10ml)
Concentration (in ppm) =	Vc
11.2
This is rearranged to give: Vc = concentration
V-22

-------
Thus to detect 100 ppm of N02 with 10 ml of absorbing reagent
without saturating the reagent, a gas sample having a volume
11.2
of approximately 112 ml as needed, or Vc = 100 = .112 liters.
The sample flask must allow an additional 10 ml for the volume
occupied by the absorbing reagent.
The effective range of this method can also be
extended by increasing the volume of absorbing reagent used in
a sample flask. For example, if 20 ml of the absorbing reagent
is saturated, 42 gms of NO£ will have reacted with the absorbing
reagent. If 112 ml of sample gas resulted in saturating these
20 ml of reagent, the NO2 concentration would be:
(.532) (42 gm/20 ml)
Concentration (in ppm) =	.112	= 200 ppm.
By varying the sample flask size (100 ml, 300ml, and
1000 ml flasks were used) and the amount of absorbing reagent
used (10 ml, 20 ml, and 30 ml were used) we were able to extend
the effective range of this method without approaching absorbing
reagent saturation. Families of calibration curves were generated
for the various flask sizes and various amounts of absorbing
reagents used (See Figures 3 and 4).
To verify the validity of this approach, several tests
were run with the burner. In test No. 218 and again in test No.
219, three NO2 samples were collected in 1000 ml flasks containing
10, 20, and 50 ml of Saltzman reagent. The results are shown in
the following table
Indicated NOx ppm
Test No.
f/a
10/1000
20/1000
50/1000
218
.0311
2.5
1.9
2.6
219
.046
19 .2
32.6
33.2
It is clear that, except for some scatter in the data, the first
run, #218, was not saturated, and gave substantially the same
result in all three flasks. Run #219, on the other hand, saturated
the 10/1000 flask, but gave comparable results for both the
20/1000 and the 50/1000 flasks.
All testing from run 218 on used adequate amounts of
Saltzman solution to avoid saturation effects.
c. Detection of NO in Gases Which are
Low in Oxygen
As mentioned in the previous section,
when the nitrogen oxides were mixed with nitrogen by collecting
over water and then analyzed for NOx virtually no color was
obtained in the Saltzman reagent. Nitric oxide is apparently
not detected by the Griess-Saltzman method but NO2 is. This agrees
V-23

-------
with Saltzman's observations (App. A). Therefore if NO cannot
be oxidized to NO2 in the flask, it will not be detected. The
lack of oxygen and the presence of water in the experiment cited
previously virtually eliminated any NO2 from the NO-NO2 mixture.
For detection of nitric oxide, Truesdail Laboratories
recommends that the gas sample and reagent be shaken in the
sampling flask for at least an hour before the dye intensity is
measured. This allows the NO to be oxidized to NO2. If there
is not sufficient oxygen in the flask, it must be added. Figure
16 shows the region where there is msufficent oxygen in an
equilibrium propane/air combustion mixture to adequately
oxidize NO to NO2- In this region oxygen must be introduced
into the sample to detect nitric oxide. The procedure used here
to achieve this was to partially evacuate the flask. This was done
by evacuating a flask fully and connecting it to another
unevacuated flask of equal size and allowing the two to equilibrate.
The sample was then drawn in this partially evacuated flask. The
oxides of nitrogen concentration level read from Figures 3 or 4
must be doubled to account for the flask being half evacuated (in
effect only half of the flask volume was used to take the
gas sample).
When these two final procedures were adopted: (1)
using smaller size sampling flasks and/or larger amounts of
absorbing reagent, and (2) introducing air (oxygen) into
samples when there is insufficient air to oxidize the NO to
NO2 the final data curve seen in Figure 18 resulted.
d. Effect of Absorbing Reagent
Temperature on Sensitivity
During the course of testing,
a question was raised regarding the storage and use of the
absorbing reagent. ASTM procedures (Nitrogen Dioxide Content
of the Atmosphere, (D 1607) pg 455 personnel at Truesdail
Laboratories, and Saltzman ( Appendix A) all contend that the
absorbing reagent will remain stable for several months if
refrigerated in well stoppered brown bottles. The ASTM
procedure further contends that the absorbing reagent should
be allowed to warm to room temperature before use. Other
workers in the field, however, feel that the absorbing
reagent should be made fresh (as close to the time of use as
possible) and should be kept cold until immediately before use.
To investigate the effect of age and temperature of the
absorbing reagent on its reactivity, a series of tests were
conducted.
Newly prepared absorbing reagent (less than 5 days
old) and older reagent (approximately one month old) were
compared at room temperature (approximately 70°F) and also
at refrigerated temperatures (approximately 35°F). Some of
the older reagent had, in fact, been stor.ed at room
temperature for approximately one month before the test. The
cold sample of older reagent had been previously stored in a
refrigerator for the same period of time. The cold new
V-24

-------
reagent had been kept refrigerated since it had been made.
The warmed new reagent was removed from the refrigerator just
long enough to come to room temperature before being tested.
Each batch of reagent was tested with the standard
calibrating solution and by introducing it into previously
sampled flasks containing the same concentration of oxides of
nitrogen. Table 2 shows the results of these tests. It is
seen that there are no significant differences between the
results from these variable conditions.
e. Effect of Evacuating Procedure
on Sampling Results
Saltzman (Appendix A), the APCD
procedure (Appendix A) and the ASTM procedure referred to on the
previous page, all indicate that an appropriate sampling
procedure is to put the absorbing reagent in the sample flask,
evacuate the sample flask to the vapor pressure of the reagent
and then obtain the gas sample. Other workers in this field believe
that the sample flask should be evacuated first, the qas sample
obtained, and then the absorbing reagent added to the flask.
This procedure is in keeping with the idea of using only fresh
cold absorbing reagent.
A test was made to investigate the effect of evacuating
the sample flask before or after the addition of the absorbing
reagent. In this case a gas sample was used to test this effect.
Table 3 shows the results of this test. Again no significant
differences are seen between these two procedures. Addition after
sampling was adopted as a standard procedure, however, because it
was generally more convenient.
2. Hydrocarbon
a. Heated Sample Line
In obtaining valid emissions data, the
FID sample line has come under close scrutiny in two respects.
First, there has been some difficulty in maintaining a clean line.
This has been a result of the unexpected introduction of hydro-
carbons into the line from valves and other component fittings. It
can also be caused by hydrocarbons brought in from a burner during
blowout testing. The second consideration has been the problem of
avoiding condensation of hydrocarbons from the sample in the line.
It has been suggested that a heated line with a
temperature on the order of 300°F would avoid this problem. The
diesel and kerosene fuel which were used in this program have a
final boiling point on the order of 550°F to 625°F as shown
in the chart of Table 4. It has an approximate molecular
weight of 170 corresponding to dodecane (Ci^H26). To avoid
condensation of the pure vapor in the line it would be
necesssary to maintain the line above this temperature. When a
gas sample consisting of several constituents is under consider-
ation, however, only the partial pressure of the hydrocarbon
V-25

-------
concerns us. This in turn depends on its volumetric concentration.
The dew point characteristic of a gas mixture containing
dodecane is shown in Figure 21. The ordinate of Figure 21 is
concentration of hydrocarbons in ppm by volume. The abcissa is
temperature in degrees F. Since the Paxve FID chromatograph is
calibrated for hydrocarbons expressed as hexane, it will read twice
the true concentrations of dodecane. This FID reading is shown
by the middle line in Figure 21.
It is seen that a sampling line at ambient temperature 70°F
will not have condensation if the indicated hydrocarbon concentra-
tion is less than approximately 1000 ppm. All of the measurements to
date on the Paxve burner in normal operation indicate hydrocarbon
readings of much less than this value. Therefore it would not
appear that condensation will contribute emission measurement
errors unless there are higher hydrocarbons present. The third
line on Figure 21 was added to indicate the relationship between
the dew point and the fuel/air ratio for diesel fuel/air mixtures.
In this case the ordinate is fuel/air ratio by weight. The top
of the graph corresponds to a fuel/air ratio equal to 1 and the
next lower decade corresponds to a fuel/air ratio 0.1 etc.
For the fuel/air range of interest (f/a = 0.04) the dew point
is 265°F. Thus a sample line heated to 300°F should permit
passage of even an unburnt mixture through the instrument without
condensation.
Although we did not expect true condensation to be a
problem with our exhaust samples, there are other phenomena to
be considered. In the curve of our FID calibration work, we have
often noted the strong tendency for hydrocarbon gases to adhere
to the walls of containers and lines. This adsorbtion phenomenon
is not unlike condensation, except that no truly liquid film need
be involved. We have found that heating the walls and lines
minimizes any adsorbtion problems. Based upon these considerations,
the sample line has been heated for all of the burner testing
data presented in this report.
b. Heated Pump
The necessity for maintaining temper-
ature control of the sample gas has been discussed. In order to
accomplish continuous sampling of the burner exhaust gas at elevated
temperature, a sampling pump was designed and fabricated by
Paxve. No commercially available pumps were found capable of
meeting the requirements. A photograph of the pump is shown in
Figure 22. It utilizes two metal bellows driven by a
rotary eccentric mechanism. The eccentric mechanism is in turn
driven by an electric motor through a long shaft. The valves in
the pump are two steel balls retained by a steel shim stock in a
valve hole with a lap seat. All parts of the pump can be operated to
over 500°F. In operation the pump is situated in a small oven
with the heated sample line coupled into the pump. The outlet
side of the pump is a heated 1/4" line which leads from the
pump through a small receiver to the flame ionization detector.
A bypass bleed is provided at the end of the FID to adjust
V-26

-------
sample line pressure. The small receiver smooths out the
pump pressure pulsations. The receiver is contained in the oven.
The FID chromatograph body is maintained at
temperature of 200°C (392°F) with the detector head about 50°
hotter. Initial operation of the Paxve sample gas pump in
connection with the FID chromatograph produced erratic results.
The chromatograph traces showed erratic fluctuations combined
with random drifting of the instrument. The erratic signals
from the line were traced to a cold spot (where calibration
gases were introduced). Condensation at this soot with
subsequent dripping of condensed water into the line was eliminated
with an immediate and permanent improvement in signal quality.
During subsequent operation of the system the line was found to be
"dirty" (high unburned hydrocarbon readings) on one occasion.
This condition was corrected by running hot exhaust products
through the line till this spurious signal was cleared.
d. Oxygen Synergism
As was previously explained in the
discussion on calibration of the FID, the addition of oxygen to the
span gas altered the linear response of the FID to hydrocarbons.
This altering of the response due to oxygen is known as oxygen
synergism. The effect of various pure hydrocarbon-oxygen mixtures
sampled in a Beckman FID using pure hydrogen is seen in Figure 23.
It is seen that the presence of oxygen, in most but not
all cases, shows a suppression of the response from the instrument.
This effect may be reduced using a hydrogen mixture with either
nitrogen or helium for the FID flame. Figures 24 and 25 show
the result. The mechanism which produces these results is not
known.
Paxve used pure H2 for our FID because we were
unaware of the synergism problem until after we had taken the bulk
of our data. A correction factor was applied to the data to account
for the reduced sensitivity of the instrument in the lean
combustion (high O2) mode. Figure IV-12 shows the oxygen
synergism effect for our FID, with our span gas.
D. References
1. Littlewood, A.B., "Gas Chromatography -
Principles, Techniques and Applications" Academic Press, 1970 .
V-27

-------
- INDICATORS
MECHANICAL ZERO
ADJUSTMENT
SCREWS
ZERO
ADJUSTER
KNOBS
OXYGEN
RANGE LEVER
CENTER
LEVER
HYDROGEN
CELL
SAMPLE
INLET
POWER CONNECTION
BAILEY HEAT PROVER
SAMPLE
FILTER
GAS
SAMPLE	OXYGEN
itl 	1.1
CHECK CELL CHECK ZEROS
1 i
COMBUSTIBLES
CELL
b.
OPERATING SCHEMATIC
FIG. 1 BAILEY HEAT PROVER WITH OPERATING SCHEMATIC

-------
COUTAC'i PIPETTE	DUBbLER PIPETTE
FIG.V-2 VOLUMETRIC GAS ANALYSIS APPARATUS

-------
OXIDES OF NITROGEN CONCENTRATION, EQUIVALENT N02,ppm
CALIBRATION CURVE FOR DETERMINING OXIDES OF NITROGEN CONCENTRATION
BY GRIESS SALTZMAN METHOD, LOW RANGE
Figure V - 3
OXIDES OF NITROGEN CONCENTRATION, EQUIVALENT NOj.ppr
CALIERATION CURVE FOR DETERMINING OXIDES OF NITROGEN CONCENTRATION
iiY GRIESS SALTZMAN I'LTtiOD, hIGh RANGE

-------
MODEL 8001 GAS CHROMATOGRAPH
USING THERMAL CONDUCTIVITY DETECTOR
TOR DETECTING CARBON DIOXIDE (C02)
AND CARBON MONOXIDE (CO)

-------
•COUUMN
NUPW3 V\£TtR\H6
THt.KM\S>"T ORb
a. THERMAL CONDUCTIVITY DETECTOR
IN»UT
" JOOV
n.

n
k

*
b. FLAME IONIZATION DETECTOR
INTERNAL CONFIGURATIONS OF GAS CHROMATOGRAPH DETECTORS
Figure
V- 6

-------
.34
.32
.30
!
.28
.26
.24
.22
100»C
Carrier Gas Flow Bate, cc/min
PEAK HEIGHT OPTIMISATION FOR CO KUITION FKOM GAS CHBOMATOGBAPH
Figure V-?
REGULATED PRESSURE, pai
CALIBRATION CURVES FOR GASES USED WITH THE FLAME
IONIZATION DETECTOR
FIGURE V 8

-------
0	400	800	1200 1600 2000	2400 2800 3000
Carbon Monoxide Concentration, ppm
m
c
o
¦H
m
300
w 200
a. Low Range Calibration
E

8
g

be
0

•p
ro
G
o
u
0

U

g
(0
CJ
g

0
6
•iH
O
CO
u
•H
Uh
>

•H

Q
JZ

W)
V
•H
H
0)
d
X
o

C/5
X.

H3



0.

Carbon Monoxide Concentration, %
b. High Range Calibration
CARBON MONOXIDE CALIBRATION ON THERMAL CONDUCTIVITY DETECTOR GAS CHROMATOGRAPH
Figure V - 9

-------
•»	6	8	10
Carbon Dioxide Concentration, %
o o
o
(0
E C
O O
M -H
Cm U.
•H
-P >
x: -h
¦H
CD 41)
X rH
* s

-------
3000
2000
1000
100
Hydrocarbon Concentration, ppm
CALIBRATION Oh' SEPARATION SIDE OK FID GAS CHROMATOGRAPH
Figure V-12
Open Points Air Dilution
Filled Points N~ Dilution

SCHEMATIC DIAGRAM OF EXPONENTIAL DILtJTICK ftPPAJUTUS
Figure V - 1?

-------
10000
w
c
o
•H
W
•H
>
Q>
%
CO
1000
jC
w
•H
0)
X
100
Time
FIC,. V-14

-------
Peak Height from Chromatogram, Scale Divisions

-------
Calibration of Flame Ionization Detector
Span Gaa Diluted with Nitrogen
x 10~3	x 10"^	x 10"^
Fraction of Full Scale Reading
From Right Side (Separation) of FID
FigUM~
V-16

-------
o
I
GJ


H ^

0 1

3 a»

o

S3 rf

H- H*

JP 0

D* 3

rt

O

CO Hf,

M-

o. ^

(D C
X


^ »-

CO
o
0> CO

T> O

QJ 01

*1 h-

0) 

rt

H.

O (D

D Q>

w a

h"

O 3

^ jp



a

< ^
I H*
h-» CQ
-o c
x 10"2
x lr--1
Fraction of Full
Scale Reading
From Left Side
(totals)of FID

-------
10,000
1,000
100
10
l.o
tiq uilibri
0 .1
Initial Data
Equilibrium NO2
0 01.
a02 0 .Ok 0 .06 0 -°8
H'uel Air Ratio, f/a
C0MPAHISON W INITIAL ANL) r'INAL DATA
Distance Between are indicated
Distance Between Burner and Sample Point #1-1 ft.
#2 - 1 ft,
#3 - 15 ft.
m - 52 ft.
NO SAMPLING POSITIONS
x
Figure V-19

-------
100
106 "|i"rO
r
s,
VP*
10
I
C MO 2/10r 1 /JiSOnjII r p^/clji
i
-------
a. Top View of Pump
b. Side View of Pump
DIAPHRAGM PUMP USED TO OBTAIN HYDROCARBON DATA
BURNER EXHAUST SAMPLING SYSTEM
Figure V-22

-------
110
a>
CO
c
a
CO
0)
0)
>
+>
H
©
a:
100
90 -
80 -

REFERANCE : BECKMAN INSTRUMENTS
<
•
N>
O
OXYGEN SYNeiHniSK iSKKKCT ON VAHI0U3 HYiJHOChHBONS injNO PiJHc; HYUKOGtsN hNiJ «1H

-------
110 i
H*
100 -*=£=
0)
CO
c
a
(0
0)

¦p
cfl
0)
3=
REFERANCE : BECKMAN INSTRUMENTS
ro
OXYGEN SYNERGISM iKFECT ON VARIOUS HYUHOC/irtBONS USING m NITHOGaJN-HYDHOGslN BL&NjJ ANiJ alR

-------
<
I
ro
cn
11C
o>
«
e
a
(0
0)
os
©
>
«J
rH
L„& 75
~4cc/^,
m
4nn Vc/mm:
l

-t- /1
• 'J "

TV

rr
1:..
t
¥
£77
J.
.. i_
-i-l-assi

, J
I	
-'4--
BUTAN0
wee—?
BurENE<-i
19


s
\%G*
-SrlfANB
PSOPANe
Hett/we |
\BiiTAPieqe
& sA +tru
J	L
4— !-
REFERANCE : BECKMAN INSTRUMENTS
OXYGtiN SYNERGISM iCKFUCT ON VARIOUS HYDHOCrtrtBONS USING k Hk.JjIhK-HYJH0Gi,;N KIaND AKu A IK

-------
SAMPLE
POINT
PROBE MATERIAL
NOx CONCENTRATION
(EQUIVALENT N02, ppm)



Test 1
Test 2
#1
(At Burner)
Ouartz
9.5
9.5
#1
(At Burner)
Stainless Steel
9.8
10.2
#3
(15' from Burner)
Stainless Steel
5.15
8.8
#•»
(52' from Burner)
Stainless Steel
0.1
2.8
COMPARISON OF NOx SAMPLING POSITIONS Table V -1

Standard Solution

Transmission at 550

(56)
New Reagent, Cold
1*2.0
New Reagent, Warmed
U-2. 0
Old Reagent, Cold
1+2.5
Old Reagent, Warmed
W.5
TEMPERATURE. ANlJ AGING EFFECTS ON GHIES3-SALTZHAN ABSORBING RKAGENT Table V-2

Transmission At
Equivalent Oxides

550 m-t<
of Nitrogen

(*)
(ppm)
Reagent added before evacuation
62.3
2.3
Reagent added after evacuation
64.0
2.2
EFFECT OF EVACUATING PROCEDURE ON SAMPLING RESULTS Table V-3

-------
FUEL SPECIFICATION - KEROSENE
PROPERTIES
Gravity, °API	39.8 (0.826)
Color, Saybolt	30+
Flash, Tag c.c., °F	136
Pour Point, °F	B-60
Viscosity, cs at 30°P	10.24
Copper Strip at 122°F	1
Copper Strip at 212°F	1
Corrosion, Silver Strip	0
Morcaptan Sulfur, '/. vt	0.0001
Odor	ok
Smoke Point, mm	20
Aromatics, V. vol	14.1
Freezing Point, ASTM °F	B-58
Water Reaction, Inc. or Dec., ml	0.5
Interface Rating	1
ASTM Distillation, °F - I.B:P.	344
107. Evaporated	365
507. Evaporated	419
907. Evaporated	4 78
957. Evaporated	491
End Point	504
Recovery, 7. vol	98
Residue, vol	1.0
Loss , 7. vol	1.0
Aniline Gravity Constant	5532
ETU/lb (Calc)	18,500
Cum, Existent, Steam Jet Mgs.	1
Cuni, Potential, Steam Jet, Mgs.	1
TAN-C	Neutral
S/VN'-C	Nil
01cfine, '/.vol	0.7
Naphthalenes, (Diaromatics)	0.31
Water Separometer Index Mod	98
Luminometer Number	4 7.3
nit.-m.il Stability. ACTM-CFR Coker
Pressure Drop, In. Hg	.05
Prehcatcr Deposit Rating	0
Table V-4

-------
VI. EXPERIMENTAL RESULTS
A. Experimental Data Listings
A complete listing of all emission and stability data
for the Paxve burner obtained during this program is given in
Tables 1 through 13. This listing consists of all the basic raw
data obtained from burner testing which was conducted intermittently
over a period from October 1970 to May 1971. The run numbers
were assigned m chronological order and are listed consecutively
with deletions of runs 173, 187, 204, 208 thru 216, 276, 277
and 278. These runs were either misnumbered or did not
apply to this program. The data tables were compiled by inputting
the test data sheets into the APL IBM/360 computer. Data sorting,
checking, editing and final printout were greatly facilitated
through the use of the computer.
All pertinent burner data that was taken is listed and
the test parameter symbols are defined in the table of nomenclature,
Table 14. Each run is categorized under test type according to
the fuel used, test objective, the burner system configuration
and the test stand in which the test was conducted. The minus
one (-1) notation given in the various columns of the data
is an IBM 360 computer expedient to indicate data elements which
are blank. The data voids will occur as a function of the
particular run objective. Where applicable, run comments are given
to describe or indicate runs which were made for procedural checks
such as stability testing techniques or emission testing techniques
to improve the quality of data. Also, in the course of the
program, as elements of the test facility were added, such as the
vapor generator loop, some runs were allocated for providing a
check of the system additions. Comments are also made to indicate
data which through the sorting technique has been indicated
to be obviously bad. Improvements were made to the burner in the
course of testing. These improvements consisted primarily of better
fuel/air mixing to eliminate fuel/air stratification. The injector
improvements were applied to the burners in both stands 1 and 2.
The comment column of the data listing indicates when the finalized
burner improvements were instituted by the designation N which stands
for new burner, in contrast to 0 for the old burner. As a further
clarification of the major events affecting the test program and
relating to the data listings, a table of significant test
program events is given in Table 15. The notes indicate by run
number when significant changes in test operation or data
evaluation techniques took place during the burner evaluation
program.
1. Explanation of Data Tabulation Column
Headings
A table of nomenclature for the symbols given
in the column headings of the Experimental Data Tables 1 thru 13 is
given in Table 14. The table of nomenclature is self explanatory
and further amplification is not necessary. The data given in

-------
the experimental data tabulation is all basic raw data with the
exception of the nominal fuel/air ratio (FAN). The detailed
explanation for the derivation of FAN is presented in Section VI-B.
The chromatographic and FID data (CO2C, NOT, NOB, CO and HC) is
given in either percent concentration or in parts per million.
These values were obtained from the measured strip chart recorder
deflections and applying the appropriate spanning factor. Span
was obtained by calibrating the instrument with a gas of known
concentration as discussed in detail elsewhere in this report.
For those runs for which the CO emissions were not
detectable, the CO values were reported as 5 ppm which was the
lower limit of resolution for the chromatograph. Those runs for
which the FID readings were below zero were reported as zero ppm.
The air flow data (WA) was taken from the meter as a
volumetric reading from Stand 1, in percent of full scale (SCM).
The meter reading was then converted to a mass flow rate by using
the meter calibration curve and correcting for temperature and
pressure in the following manner:
/Tind°R
(wa)corr. = Pstd.  VTSTDOR) PSTD (PSIaJ
Stand 2 operated slightly above 1 ATM and both pressure and
temperature corrections were applied to the data.
The fuel flow readings (WF) for propane were read in
standard cubic feet per hour for both Stands 1 and 2. These
meter readings were converted to mass flow in pounds per hour
as a function of the propane supply pressure by calibration curves.
Temperature variations at the meter were negligible and
corrections were not necessary. The kerosene liquid flow meter
was nearly linear over the required flow range and the meter
readings were converted to mass flow rate in pounds per hour
by referring to the calibration curve. Corrections for pressure
and temperature were unnecessary since liquid flow at the flow
meter was maintained at essentially ambient temperature conditions.
VI-2

-------
As described elsewhere in this report, it was found that
the FID chromatograph was sensitive to oxygen concentration. A
correction factor has been applied to all the hydrocarbon emission
measurements given m Tables 1 through 13 in accordance wi,th the
calibration factor given in Figure IV-12 when oxygen was present
in the gas sample.
2. Explanation of Significant Test Program
Events
In the course of the burner evaluation program,
significant events occurred which influenced the quality of the
data. Facility improvements were added and procedural methods were
developed which aimed at obtaining data of the best quality. The
more important events in this regard have been listed to provide
the reader with a better understanding of the progression in
testing and instrumentation technology achieved during the course
of the program.
a.	Stability Test Procedure
The test program began with an
evaluation of burner stability lean limit with propane on Stand 1.
Runs 1 through 7 were devoted to exploring the best indicators for a
true indication of the lean blowout limit. These tests
established a procedure which was utilized in subsequent runs
and this group of runs in themselves did not produce valid
stability data. The initial difficulty in establishing the lean
blowout limit was associated with the fact that the residual heat
capacity of the burner appeared to sustain burning beyond the lean
limit fuel/air ratio. It was found necessary to make a number
of runs on and near the lean limit point for extended time duration
to establish that the burner remained at a steady state burning
condition.
b.	NOx Line Loss Check Out
During initial testing the gas samples
for NOx analysis were drawn from the long sample line which provided
the gas samples for all other emission data. The readings became
suspect from this testing configuration when only very low values
were consistently obtained. Beginning with run 95, a series of
tests were conducted in both test stands 1 and 2 to evaluate the
effects of drawing the NOx sample from various line lengths and
from sample lines of different materials. It was found that NO2
was being absorbed by the condensed water vapor in the long
sample line and a procedure was established for drawing NOx sample
through short quartz tubes which gave satisfactory results and
was followed for the remaining runs.
c.	Propane Accumulator Installed
Stand 2
It was noted during testing that
propane flow oscillations were occurring in test stand 2. Due to
low propane supply pressure, pressure coupling with the burner
VI-3

-------
chamber pressure oscillations occurred which caused oscillations
in flow and interferred with the propane flow measurements. This
difficulty was eliminated by installing a propane accumulator
downstream of the propane supply tank. This change was made during
run 10 8 and served for the balance of the Stand 2 testing m the
program.
d.	Air Flow Straightener Stand 2
During initial testing in Stand 2 it
was noted that unsymmetrical air flow profiles existed at the
burner inlet. The air flow plumbing in stand 2 for which the vapor
generator loop is utilized, requires that a 90° bend be installed
upstream of the burner inlet. To eliminate the effects of the flow
discontinuity, a flow straightener consisting of a bundle of small
diameter tubes was installed at a secion downstream of the 90°
bend which eliminated the problem.
e.	Hot Sample Line Installation
During initial emissions testing,
long unheated sample lines were used to draw the sample to the
chromatograph instruments. It was suspected that condensation of
some of the exhaust gas constituents could occur in the sample line.
This is particularly true of the unburned hydrocarbon fuels or
decomposition products which might exist in the exhaust gas
sample. Difficulty was also encountered in hydrocarbons or
impurities being retained in the long sample line from previous
running modes and obscuring the data for following runs. This
problem was resolved by installing a line heater over the entire
length of the sample line and maintaining the line temperature at
300°F or above. This temperature was sufficient to vaporize any
of the heavier fuel fractions occurring in kerosene. The hot
sample line was installed and checked out during run 128 for Stand
2.
It was necessary to complete the sample line heating
system by installing a specially designed and fabricated
electrically heated diaphram pump in the sample line. A pump was
required to maintain the necessary pressure for accurate
chromatograph data. The pump installation was completed and
checked out at run 138 with completely satisfactory results.
From this point on the heated system was used in all burner
testing.
f.	No Data Runs
Run Nos. 173, 187, and 204 are runs
during which no data was collected due to misnumbering of the run
sequence or due to transient variations which did not permit
collecting steady state data. Run groups 208 through 216 and
276 through 278 were made as special runs using special burner
equipment for other purposes and are not pertinent to this
program.
VI-4

-------
g. Liquid Kerosene Runs
Runs 203, 342, and 348 through 350
were made with liquid kerosene instead of vaporized kerosene. The
liquid kerosene was atomized by injecting a small amount of
nitrogen into the fuel line. These runs are identified to
differentiate the normal kerosne runs which were conducted with
vaporized fuel.
h.	Vapor Generator Loop Check Out
A vapor generator system was installed
in Stand 2 to operate in conjunction with the burner. The purpose
of this configuration was to study the effect on emissions of
exhaust gas quenching over the vapor generator heat exchanger. This
system consisted of a complex arrangement of mechanical,
hydraulic and pneumatic components as well as an engine driven
pump. It was necessary to run the burner in conjunction with the
vapor generator loop to establish th effect of thermal conditions
on the operation of the loop. This was accomplished during
runs 205 through 20 7.
i.	Vapor Generator Stack Clean-Out
During the early emissions testing it
was found that higher hydrocarbon concentrations were being emitted
from the top of the vapor generator stack than from the burner at
the bottom of the stack. In the course of investigating the
possibility of hydrocarbon emissions being generated from
accumulations on the vapor generator coil, a special test was
run in which the vapor generator coil was exposed to the burner
exhaust gas while flowing nitrogen through the loop instead of
the normal working fluid. This procedure was done in order to raise
the coil temDerature sufficiently to vaporize any absorbed
hydrocarbons.
j. NOx Saturation Evaluation
Further doubt was cast upon the NOx
gas analysis procedures which were being used by the fact that
limiting values for NOx were being obtained which never seemed to
be exceeded for the various runs. It was suspected that the
proportions of Saltzman solution per sample which was being
used for colormetric analysis might be insufficent to
avoid saturating the solution. To investigate this possibility
further three one liter flasks were prepared containing 10, 20, and
50 ml of Saltzman solution respectively.
During test No. 218, burner exhaust samples were
collected in each of the three flasks while the run condition
was maintained constant. It was found that the 10 ml flask
indicated a maximum NOx value of 19.7 ppm, the 20 ml flask
indicated 33.2 ppm, and the 50 ml flask indicated 32.6 ppm.
These data indicate that saturation of the Saltzman
solution was indeed occurring. The NOx analysis technique was
modified as described in detail elsewhere and checked out fully
VI-5

-------
during runs 219 and 220. This technique was then utilized for
all subsequent runs and gave satisfactory results.
k. Fuel Injector Improvements
During runs 220 - 2 37 it was found
that hydrocarbon readings from the top of the vapor generator stack
were much higher than those obtained from the burner. A series
of investigations were conducted to determine the cause of this
phenomenon. Oil deposits on the vapor generator coils and
leakage of organic working fluid from the vapor generator tubing
were both investigated and ruled out as possible explanations for
the high HC readings. The fuel distribution patterns m the
burner were then examined. The oxygen, carbon dioxide and
hydrocarbon levels were determined at various locations
in the plane of the burner exhaust. It was found that the exhaust
flow was highly non-uniform. Disassembly of the burner revealed
a damaged fuel injector which was flowing most of the fuel into
one side of the burner. To remedy this situation and promote
even better mixing, new fuel injectors were introduced into
both the Stand 1 and Stand 2 test installations. This work was
partially documented in runs 279 - 281. During this time, the
joint between the inlet pipe and the burner was also
modified to prevent raw fuel from being carried up the stack
without passing through the burner.
After the modifications noted above had been made to the
test installations, it was found that hydrocarbons data from the
burner and the top of the vapor generator stack were in substantial
agreement. It was also found in later examination of the data,
that the NOx readings from the burner were improved as compared to
data taken before the modifications. Data taken during and
subsequent to run 282 reflect these improved results.
B. Fuel Air Ratio Analysis and Correlation
1.	Introduction
The ratio of fuel mass flow to air mass flow
is a very important and basic correlation parameter in burner
performance evaluation. All emission and stability characteristics
of the burner are established on the basis of operation at a given
fuel/air ratio. To provide the most accurate definition of burner
emission and stability performance and to reduce data scatter,
considerable effort was made to establish a reliable fuel/air ratio
for each test run. Various sources of measurement were employed to
establish a true value of fuel/air ratio for each run. The
following discussion will describe the various methods used and
the rationale in selecting the value of fuel/air ratio which was
assigned to the particular run.
2.	Methods of Fuel/Air Ratio Measurement
Values for fuel/air ratio were obtained for
each run from various instrumentation sources using two methods
of determination. The first method employed the direct flowmeter
VI-6

-------
measurement of fuel flow and air flow. The appropriate calibration
factors to the flow meters were then applied and corrections were
made for temperature and pressure to the fluid flows in arriving
at a test fuel/air ratio. The second method employed an indirect
means of determining the fuel/air ratio by measurements and
analysis of the exhaust gas. The indirect method employed two
instruments to measure the exhaust gas oxygen concentration and
carbon dioxide concentration as well as the unburned combustible
constituent concentration. These measured constituent concentra-
tions were then compared with the theoretical equilibrium compo-
sition for the combustion of hydrocarbons as given m Figure 1
for the given fuel/air ratio. The exhaust gas carbon dioxide and
oxygen composition was measured by the volumetric (Orsat)
apparatus during steady state combustion operation. In
addition, the Bailey meter gave continuous readings of
oxygen composition and combustible gas composition of the exhaust
gas sample. The Bailey Heat Prover was used during the burner
testing to estimate the burner operating point. The Heat Prover
has two meters, one which reads oxygen, and the other, labeled
combustibles, which reads a mixture of the hydrogen and carbon
monoxide present. For lean runs, the oxygen meter was used in
conjunction with theoretical exhaust composition curves to find
the approximate fuel/air ratio at which the burner was operating.
For rich operation the Bailey combustibles meter was used in
conjunction with an especially prepared curve (see Fig. 1) for
the same purpose. Figure 2 shows the comparsion of the Bailey
oxygen data with the volumetric oxygen information. Although
there is fair agreement, it is clear that the Bailey reads low
at the higher oxygen values. As a further check, the chromatograph
(TC) was used to also measure CO2 and CO for a comparative
examination of the operation fuel/air ratio. The specific instru-
ments used in both methods of fuel/air ratio determination are
discussed in extensive detail in other secions of the report and
will not be discussed further here.
3. Assessment of the Methods of Fuel/Air
Measurement
In the analysis of fuel/air ratio measurement
accuracy the fuel/air ratios indicated by the various instruments
were tabulated for all the runs as shown in tables 16 thru 23. The
basic measurements which are given in the tables are:
a.	FA, which is the fuel/air ratio as measured
by the flow meters.
b.	FAO, the fuel/air ratio determined by the
oxygen reading from the Orsat apparatus.
c.	FAC, the fuel/air ratio as determined by
the carbon dioxide readings of the Orsat.
d.	FAB, the fuel/air ratio determined by the
Bailey apparatus.
Of the three basic measurements the greatest weight was given to
VI-7

-------
the volumetric Orsat analysis. This instrument has the best
inherent accuracy (approximately 0.1% of reading) and further
provides a self contained check in the balance between the oxygen
and the carbon dioxide measurements. The Bailey meter is
convenient as a continuous recording device but is less accurate
than the Orsat and was used primarily for a check of the Orsat
readings in the lean combustion mode. As will be noted from
the data of Table 24, the flow meter fuel/air ratio values
during the early phases of testing in Stand 1 and in Stand 2
fall below the Orsat readings. This fact is attributed to
air leaks tht existed in the facility plumbing downstream from the
flowmeter which gave erroneously high air flow readings. This
would result in an indicated lower fuel/air ratio at the burner than
true value. The lower fuel/air ratio readings from the flow
meters as compared to the volumetric were noted in particular
while testng in Stand 1 during rhe early test runs. During a
later test period an overhaul was made of the Stand 1 air plumbing
to eliminate all possible leakage points by welding all joints and
it is noted that the correlation with volumetric data improved
markedly. The consistency of the Orsat volumetric readings of
fuel/air ratio derived from carbon dioxide and oxygen exhaust gas
concentration measurements are shown in Figure 3. The general
consistency of the volumetric measurements and the greater
inherent accuracy of the apparatus makes it a first choice or
primary standard in establishing the true value of fuel/air ratio.
It is further noted that the volumetric measurements of fuel/air
ratio are applicable with the greatest degree of confidence in
the regimes of combustion where the equilibrium composition
can be reasonably established as given in Figure 1. The Orsat
fuel/air ratio determination therefore is not applied at the lean
limit or rich limit points.
In the early phases of testing, Orsat data was taken
periodically and is not available for every run. This was done
to expedite testing since the Orsat analysis requires that a
grab sample be taken and a rather time consuming process is
required in performing the gas constituent analysis during the test
run. Since the Orsat data is not available for every run, it was
necessary to arrive at an adjusted composite of fuel/air data
which could be applied for every run to arrive at the true
fuel/air ratio. In order to arrive at cross-correlation of
fuel/air ratio from the various instrumentation sources the method
described below was employed.
4. Description of Fuel/Air Ratio Correction
Procedure
A tabulation of all the fuel/air ratios derived
from the various measurements is given in Table 16-2 3. The FA
column is a listing of the fuel/air ratios obtained from the flow
meters and is uncorrected. The column headed FA COR is the flow
meter fuel/air ratio determination as corrected by factors which
will be described. Correction factors for the flow meter data
were determined on the following basis.
VI-8

-------
a.	Propane, Stand 1 - early tests
b.	Propane, Stand 2 - later tests (after air
plumbing leak repair)
c.	Propane, Stand 2
d.	Kerosene, Stand 2
The correction to the fuel flow meter readings is derived from the
Orsat data using selected runs for which the Orsat data was
available. The above correction factors were determined for
lean combustion, maximum burner efficiency operation for
which the theoretical equilibrium combustion criiteria given in
Fig. 1 is closely applicable.
The flow meter fuel/air value was divided by the value
obtained from the average of the Orsat CO2 and O2 fuel/air determin-
ation. The ratio thus determined for each of the selected runs was
averaged by the computer for all the particular selected run
group and a standard deviation was determined. Obviously bad
data points which deviated in excess of three standard deviations
were excluded. A final correction factor was then obtained by
re-averaging all of the ratios. In this manner, correction
factors were obtained for the propane stand 1 early testing and
later testing, the propane Stand 2 testing, and the kerosene Stand
2 testing. A tabulation of the flow meter correction factors is
shown in Table 24. These correction factors are to be applied to
the appropriate values of FA to arrive at the values listed under
FA COR. Referring again to Tables 16 thru 23, the column heading
given by FAO is made up of fuel/air values obtained from the
Orsat volumetric oxygen measurements. The fuel/air values under
FAC are the corresponding Orsat values derived from the CO2
measurement. The values for FAB are the fuel/air ratios obtained
from the Bailey Heat Prover. The column headed .FAN gives nominal
fuel/air ratios upon which the reported burner performance is based.
As a result of the previous discussion and general rationale, the
FAN values are based primarily on the Orsat measurements. The
corrected flow meter data (FACOR) was used for test runs in which
the volumetric data was not available and at lean or rich limit
stability test points where combustion efficiency is reduced
and non-equilibrium values of CO2 and O2 are generated. The Bailey
Heat Prover fuel/air measurements (FAB), as derived from the
measured concentration of combustible consitituents, are also
applied for rich mode operation when the Orsat data was not
available.
5. Cross Correlation of Fuel/Air Measurements
Further cross correlation among the various fuel/
air instrumentation sources are shown in the following plots.
Figure 3 shows a comparison of the fuel/air ratios derived from the
measured Orsat oxygen concentration versus the corresponding Orsat
CO2 concentration. It is seen that excellent correlation is
obtained on an x = y theoretical line. Figure 4 shows a comparsion
of the nominal fuel/ air ratio (FAN) versus the C02 concentration as
measured by the Orsat. The theoretical equilibrium CO2 concentra-
tion is also shown superimposed on the data and it is seen that very
good correlation is obtained with minimal data scatter over the
VI-9

-------
entire range of fuel/air ratio values from lean to rich operation.
Figure 5 is a similar comparsion of the nominal fuel air ratio (FAN)
versus the O2 concentration as measured by the Orsat. This curve
also shows the theoretical equilibrium oxygen line. Here again, it
is seen that the nominal fuel/air ratio (FAN) provides a good basic
correlation between the measured O2 concentration and the
theoretical for combustion of propane and kerosene. When the
corrected fuel/air ratio from the flow meters (FACOR) is used as a
reference to plot CO2 concentration (from Orsat) as shown in Figure
6, it is seen that the data scatter increases. When the uncorrected
values of fuel/air ratios from the flowmeters are examined in a
similar manner, as shown in Figure 7, it is again apparent that
much wider data scatter is in evidence. A comparsion of the
fuel/air ratio derived from the Bailey Heat Prover versus the
Orsat fuel/air ratio from the oxygen measurements is shown in
Figure 2. It is seen here that reasonable correlation is obtained.
It appears that the Bailey has a systematic error and reads O2
concentrations which are slightly lower than the Orsat values.
The Bailey combustibles, however, gave values of fuel/air which
were in good agreement with Orsat CO2 values for rich operation.
On the basis of the above comparsions of fuel/air
ratio measurements, it has been shown that the selection of the
nominal fuel/air ratio (FAN) gives the closest correlation of
hydrocarbon combustion characteristics with the theoretical values.
This selection of a nominal fuel/air ratio will best serve as a
correleating parameter m the documentation of the Paxve burner
emissions and stability characteristics.
C. Experimental Emissions Data
1. Carbon Dioxide Data
Carbon dioxide is not an objectionable emission
resulting from hydrocarbon combustion. Its measurement during this
test program serves primarily to provide an index of the
completeness of combustion and to establish a basis for the
determination of fuel/air ratio.
Figure 4 shows all of the volumetric data for carbon
dioxide plotted against the nominal fuel/air ratio (FAN). The
good agreement is of course a result of the method by which the
nominal fuel/air ratio was selected, which is discussed in detail
in a preceding paragraph. In general, the averaged CO2 and O2
volumetric data was used when possible for the nominal fuel/air
ratio. When this could not be done the next choice was the CO2
data, then the Bailey combustibles data, and finally the corrected
fuel/air ratio from the flow meters. The fuel/air ratio as
measured from the flow meters was corrected to give the best
average agreement with the volumetric data.
Figure 4 shows the CO2 data as measured from the burner
by the volumetric apparatus. Figure 8 shows the CO2 data as
determined by the chromatograph measured from the top of the vapor
generator loop. The comparsion of these two values is shown in
Figure 9. Computer evaluation of the ratio of CO2C to CO2V was
VI-10

-------
conducted. It was determined that the average ratio of these
two values was 0.993 +0.120.
2.	02 Data
The measurements of oxygen in the burner
exhaust gas products were taken, in general for the same purpose
as those for carbon dioxide mentioned above.
Figure 5 shows the volumetric oxygen data from the
burner plotted against the nominal fuel/air ratio (FAN). As
discussed previously the method of selecting FAN causes this
curve to lie close to the theoretical oxygen curves shown in the
figure. Chromatograph data for the oxygen concentration was not
reduced since data from various other sources were available.
As noted previously the volumetric oxygen data is in
close agreement with the expected values based on the volumetric
carbon dioxide data. Figure 3 shows the fuel/air based on the
volumetric oxygen compared to the fuel/air based on the volumetric
CC>2. A computer evaluation of the average ratio shows the value
to be 1.0174 +0.0374.
3.	Carbon Monoxide Data
Figure 10 shows all of the carbon monoxide
data obtained from the burner plotted versus the nominal fuel/air
ratio. The interpretation of this data will be discussed in more
detail in Section VII. It will be noticed however, that for
fuel/air ratios above approximately 0.04, carbon monoxide data
agrees generally with the theoretical prediction based on the
information presented in Purdue University Bulletin (Ref. 1). The
higher values of carbon monoxide obtained at low fuel/air ratios
near the lean stability limit has been attributed to inefficient
combustion. Correlation of this data is discussed in Section VII.
Figure 11 shows the carbon monoxide data measured from the
top of the vapor generator stack. A comparison of this data with
the values obtained from the burner shows that there is substantial
agreement for the high fuel/air ratios. Figure 12 shows a
comparison of carbon monoxide data taken from the bottom of the
stack (the burner) and the top of the stack on the same run.
Examination of this Figure indicates that the top of the stack
values of CO are lower than those at the bottom of the stack. This
is attributed to the oxidation of carbon monoxide in the gas flowing
through the stack. In general, the maximum values obtained
at the top of the stack are lower for the same fuel/air ratio than
those obtained at the bottom of the stack.
4.	Hydrocarbon Data
Figure 13 shows all of the hydrocarbon data
obtained from the burner. The hydrocarbon data is generally
characterized by low readings over a wide range of fuel/air
ratios near the so-called operating point of the Paxve burner.
As lean blow out is approached hydrocarbon values increase for
VI-11

-------
kerosene becoming rather large near lean blowout. This phenomena
is not observed in general for propane operation. As lean blowout
is approached the hydrocarbon values remain substantially zero
until the burner is actually at the lean limit. This fact was
useful in establishing the lean limit during burner stability
investigations. As the lean blowout limit was aporoached, the
flame ionization detector measuring the hydrocarbon output was
always the first instrument to register incipient lean
blowout. When the FID first showed an increase in signal, the
operator would stabilize the burner at that point and wait. In
general, in a matter of 10 to 15 minutes, the other burner
operation indicators such as the temperature and the oxygen
indication on the Bailey would have begun to rise and if allowed
to continue, burner blow out would ensue. The operator could always
prevent this lean blowout condition from occurring by a slight
increase in fuel/air ratio which would cause the FID to once
again drop to the low value characteristic of normal burner
operation.
The use of the FID as a blowout indicator was not
found to be effective for kerosene runs. Kerosene operation as
with propane operation generally showed zero or slightly
negative readings on the FID during normal burner operation.
However, as the lean limit was approached with kerosene
positive readings of hydrocarbon, content began to appear and
would stabilize at some measurable value. This was in contrast
to operation with propane where either the hydrocarbon reading
was essentially zero or it would not stabilize but would
continue to increase with time and the burner went out. A
further decrease in fuel/air ratio for the kerosene operation caused
a further increase in the hydrocarbon output. This continued until
incipient blowout was reached at which point the hydrocarbon
signal would no longer stabilize but would continue to climb
as the burner flamed out.
An examination of the hydrocarbon data near stoichiometric
or rich operation of the burner shows that as with the carbon
monoxide there is a gradual increase in hydrocarbon content in the
exhaust with increasing fuel/air ratio. The increase in hydro-
carbons starts as approximately 90% of stoichiometric for
propane and slightly lower values for kerosene. We see again that
the kerosene provides higher hydrocarbon readings for a given
equivalence ratio than propane during rich operation.
Figure 14 shows hydrocarbon data measured from the top
of the vapor generator stack. We see here that almost all of the
runs are plotted as zero hydrocarbons. In fact, most of these runs
indicate negative readings but as discussed previously the negative
values must be interpreted as being a result of zero shift of
the flame ionization detector due to the presence of water vapor.
The low hydrocarbon readings evidenced by the plot of the
top of the stack data should not be interpreted as a difference
between the top and bottom of the stack. In fact, lower
values were obtained at the bottom of the stack than the top.
The main reason for the great accumulation of zero and very low
VI-12

-------
hydrocarbon readings in this Figure is that the fuel/air range
of operation was fairly narrow. When the burner was being
operated in conjunction with the vapor generator it was not
desirable to use a high value of fuel/air ratio since this would
create high flame temperatures which in turn would lead to
degradation of the working fluid and difficulty in operating
the vapor generator loop over extended time periods. Low
values of fuel/air ratio approaching lean blowout were avoided
as much as possible during loop operation, particularly when we were
operating with kerosene. The reason for this is that the vapor
generator coils tend to act like a trap for hydrocarbons and
lean operation with significant quantities of hydrocarbon
emissions would deposit some hydrocarbon materials on the coils
which then continue to flow into the flame ionization detector on a
subsequent run. On one or two occasions an inadvertant lean blowout
occurred due to operator error. When this situation arose during
kerosene testing, it was necessary to conduct sustained operation
of the vapor generator loop with clean burner exhaust to remove all
traces of hydrocarbon from the loop coils and the FID output.
One of the notes referring to the run data tabulated in
Tables 1 thru 13 referes to the use of a new injector. Prior to
run 282 we found high levels of hydrocarbons in the vapor
generator exhaust although the burner exhaust showed low or even
negative hydrocarbon readings. We were prepared to believe that
the hydrocarbons might disappear on the way up the stack but we
found it difficult to accept the idea that they would be generated
in the stack and hence we conducted a series of investigations
designed to explain this anamolous behavior.
Several causes for spurious hydrocarbon signals were
considered. The first of these was the acutal emission of some
hydrocarbon material that had existed in the fuel and air
intake pipe but failed to move into the burner. Such a situation
could arise due to improper seating of the fuel and air intake
pipe at the burner mouth. It was found that, in fact, there was
leakage of raw fuel/air mixture up the vapor generator stack and
this situation was corrected by providing an adequate seal at
the joint between the burner inlet pipe and the burner. This
problem is one which is peculiar to the test installation under
study here and not a problem which should be considered a factor
in burner development for automotive application.
Another factor which was considered a possible source
of hydrocarbons in the stack was leakage of the organic working
fluid from the loop. Extensive investigation of possible
sources of leakage in the loop were conducted and none were
located. This included pressure checking and helium leak
testing. All welds were found to be sound and fittings were
tight.
The third possible source of hydrocarbons in the loop was
oil deposited on the walls of the vapor generator stack or the coils
of the loop. The stack was cleaned out by operating the burner with
the coil removed so that the exhaust gas exiting from the burner was
still at high temperature. This did in fact reduce some of the
VI-13

-------
hydrocarbon levels during subsequent tests. The coil was cleaned
by passing dry nitrogen through the inside of it while it was
held in place and subjected on the outside to the exhaust gas
from the burner. This further reduced some of the spurious hydro-
carbon signals. Nevertheless hydrocarbon readings continued to
show on the order of 30 to 40 ppm from the top of the stack while
no comparable values could be found from the burner. The fuel
injector assembly was then disassembled and its flow pattern
observed. It was found that a serious maldistribution of
flow existed in the fuel injector. This caused a non-uniform
fuel/air distribution to the burner with locally excessively
rich zones. Subsequent modification of the injector succeeded
in eliminating the maldistribution of fuel. Tests after run
282 were conducted with a new fuel injector configuration which
assured adequate mixing of the fuel and air in a homogeneous fashion
prior to entrance of fuel/air mixture into the burner.
Although the burner is successful in eliminating
hydrocarbon emissions over a range of fuel/air operating
conditions, it nevertheless emits hydrocarbon under two sets of
conditions:
a.	operation near lean blowout with kerosene
b.	extremely rich operation near blowout with both
kerosene and propane.
When a poorly mixed fuel/air mixture passes through the burner,
local regions of the flow can be near rich or lean blowout while
other portions have normal good operating fuel/air values. If the
probe which is sampling the burner is in the portion of the burner
exhaust which is at the nominal fuel/air ratio no emissions will
be seen here. However, the averaged values finally exhausting from
the top of the stack after traversing the long mixing length may
contain properly burnt and improperly burnt material which issued
from the burner. Thus if a portion of the burner is very rich,
while the rest of it, including the probe location, is at a normal
fuel/air ratio, we can expect to see apparently clean operation of
the burner accompanied by excessive hydrocarbons from the top of the
stack. Improvement in the fuel injection pattern eliminated this
problem and provided a wide range of hydrocarbon free operation
on top of the stack comparable to that previously measured from
the bottom of the stack.
5. Oxides of Nitrogen
Oxides of nitrogen (NO2 and NO) formed during combustion
of hydrocarbon fuels with air are among the more objectionable
atmospheric pollutants. One of the primary objectives of this
test program was to document over a wide range of conditions ,
the oxides of nitrogen emissions from the Paxve burner. This
data is given in the curves shown in Figures 15 through 19
representing the complete nitrogen oxides emissions documentation
of the burner during the test program. Figure 15 shows the NOx
emissions measured from the burner alone while Figure 16 shows
the NOx emissions from the burner with the vapor generator
installed. Both of these plots show NOx emissions as a function of
Vl-14

-------
nominal fuel/air ratio. These data show no appreciable
difference beyond the normal data scatter. Figure 17, shows
the oxides of nitroqen measured at the top of the vapor generator
stack as as function of nominal fuel/air ratio. Here again there
is no distinguishable difference in the NOx emissions between
the burner and the top of the vapor generator stack. Figure 18
presents a comparison of emissions data between the top and the
bottom of the vapor generator stack. This comparison shows that
the emissions from the top of the vapor generator stack are
essentially unchanged from those measured at the burner. It is
also noted that there is no distinguishable influence on NOx
emissions by the inlet temperature conditions or the air mass flow.
Referring again to Figure 15, we wish to point out that the bulk
of the emission data was taken on the lean side of stoichiometric.
In observing the variation of the composite data with fuel/air
ratio as shown in Figure 15, it can be seen that the NOx emissions
increase with increasing fuel/air ratio to a maximum occuring
at or near stoichiometric. This is consistent with the data
shown in Figure 19 which shows NOx as a function of combustion gas
temperature. It is seen that NOx emissions increase directly
with gas temperature. Since the combustion gas temperature
reaches a maximum near the stoichiometric point one would expect
that the highest emission rate of oxides of nitrogen would occur
as shown in the data of Figure 15. Over the normal operating range
of the Paxve burner which is nominally at a fuel/air ratio of
0.034 the NOx emissions are maintained within a ranqe of 10 to 20
ppm. All of the data in Figures 15 thru 19 was obtained using the
finalized oxides of nitrogen analysis collection and gas analysis
techniques as described during previous sections.
D. Experimental Stability Data
Figures 20 through 26 show experimental stability
data measured on the Paxve burner. Early stability data was
taken in Stand 1 using propane together with a burner of
relatively small internal volume (33 cu. in.). Later data on a
considerably larger burner was obtained in Stand 1. This larger
burner had a volume of approximately 66.5 cu. in. All of the
kerosene experimental data was obtained in test stand 2 on a
burner having an internal volume of 52.3 cu. in. Tests were
conducted with kerosene for lean blowout only due to limitations
in the test equipment. The parameters varied during the stability
testing included the flow rate, the mixture ratio, and both the air
and fuel temperatures independently.
Figure 20 shows the stability data obtained with propane
on test stand 1 using ambient air and fuel. The circles represent
runs for which stable operation of the burner was obtained. The
triangles represent runs which were considered to be lean limit
operations. Lean limit points are those in which either (1) combus-
tion was sustained but was erratic in operation with incipient local
flameout due to minor fluctuations in air flow, or (2) the burner
did go out after a long period of time (on the order of 10
minutes or more). Squares represent runs for which the burner was
below the lean limit. In this mode it continued to operate for
a short period of time while decreasing in temperature. Flame
VI-15

-------
out would then occur unless the fuel/air was adjusted to a
condition of stable operation.
Superimposed on Figure 20 is the prediction of
stability limit at ambient operating conditions. That prediction
was made using the burner theory discussed in Section VII.
Examination of Figure 20 shows that the theory and the experimental
correlation agree remarkable well at high flow rates. At low flow
rates the theory predicts stable operation will be possible under
leaner conditions than those for which stable operation was actually
achieved. This disparity between the theory and the experiment at
low flows is attributable to the fact that the theory does not take
into account heat loss from the burner. At low flows this heat
loss can be a significant factor and can be altered by appropriate
thermal design of the burner. For the Paxve burner tested, there
was no attempt made to reduce radiation heat losses from the
burner. The heat rejection by radiation constitutes the major
thermal loss influencing lean limit operation.
The lean stability analysis given in Section VII
required the calculation of adiabatic flame temperature for each
run condition or fuel air ratio and burner inlet air temperature.
This calculation was accomplished by writing a computer program
designated CAL for use with the APL IBM 360 computer. The results
of the calculations are given in Tables 25 thru 32. Theoretical
flame temperatures are given for each run of the test program. The
temperature calculation considers air inlet temperature and fuel
characteristics.
E. Detailed Emissions Investigation
As a means of examining in further detail the influence of
air flow, inlet temperature, and vapor generator loop on critical
burner emissions, expanded plots of the data were made. Data
segregation took into account the various burner volumes which were
tested and improvements in data quality derived from improvements
to the experimental techniques relating to the measurement of
oxides of nitrogen and hydrocarbons. In addition, the data plots
show the influence of the injector modifications which were made to
improve fuel/air mixing and distribution. The data taken from run
282 on was obtained using the modified injector configuration and
includes all of the latest experimental measurement techniques for
MO and HC. This data is considered to be the most representative of
the emissions characteristics of the Paxve burner. The emissions
data is given in parts per million as well as grams/kilogram of
f uel.
1. Carbon Monoxide Emissions
Figures 27 through 37 show carbon monoxide
emissions in ppm correlated against nominal fuel/air ratio. Figures
38 thru 48 are the corresponding plots converted to carbon monoxide
concentration in grams per kilogram of fuel. The points are
identified for air flow rate, run data from run 282 on, and
inlet air temperature.
VI-16

-------
2. Unburned Hydrocarbon Emissions
The unburned hydrocarbons plotted against
nominal fuel/air ratio in ppm are shown in Figure 49 thru 55.
These data indicate the effects of lean operation down to the
blowout limit on unburned fuel emissions. The effect on rich
operation is also shown. Corresponding plots of hydrocarbon
emission concentration in grams/kilogram of fuel are shown in
Figures 56 thru 64. The hydrocarbon data shown was all taken
after the heated sample lines and heated sample line pump
were installed. In addition, the runs after modification of the
fuel injector (from run 282 on) are identified. The hydrocarbon
data shown includes both the measurements taken at the burner and
those taken from the top of the vapor generator loop.
3.	Oxides of Nitrogen From the Burner
NOx emissions from the burner plotted against
nominal fuel/air ratio for concentrations m both ppm and gr/Kg
of fuel are shown in Figures 65 thru 74. This data was all taken
after the initiation of the short quartz tube sampling line
technique and the non-saturation precautions for the Saltzman
reagent. The influence of nonstratified fuel injection and optimal
mixing of fuel is shown in the data from run 282 on.
4.	Oxides of Nitrogen from the Vapor Generator
Stack
Similiar plots to the preceding for NOx taken
from the top of the vapor generator stack are shown in Figures 75
thru 82. These data, are of course limited to runs taken with the
vapor generator installed and all these data were taken in test
stand 2. The comments relative to the sampling and gas analysis
techniques applied to the burner NOx emissions also apply for the
vapor generator NOx emissions measurements.
VI-17

-------
EXPERIMENTAL DATA FROM THE
UN
TEST
T/l
TP
V/l
WF
CO 2 7
C02C
02V
02B
NOT
W .
TIPE
of
«F LB/HR LB/HR
PCT
PCT
PCT
PCT
PPM
1
PSB1
75
71
22 . 3
0 .54
"l. 0
~1. 0
"1.0
18.7
0
•
1
2
PSB1
80
71
22.2
0.61
~1. 0
"1.0
"l. 0
11 .0
"1.0
3
PSB1
65
72
45.3
1.13
~1. 0
"l. 0
"1.0
9.5
"l. 0
4
PSB 1
85
73
45.3
1 .06
~1. 0
"1.0
"l. 0
10 . 0
"1.0
5
PSB 1
85,
73
47 . 0
1.13
1.0
1.0
"l. 0
1. 0
1.0
6
PSB 1
88
74
61 .8
1 .44
"l. 0
~1. 0
"l. 0
10 . 5
"1.0
7
PSB 1
90
75
61.7
1.58
3 . 8
~1. 0
14.0
8.5
1.0
8
PSB 1
90
73
61.7
2 .02
*1.0
"1.0
~1. 0
"l. 0
"1.0
9
PSB 1
90
74
61.7
1.62
"1.0
"1.0
~1. 0
"l. 0
"1.0
10
PSB1
90
74
61.7
1.50
"l. 0
"1.0
~1. 0
"1.0
"l. 0
11
PSB 1
90
74
61.7
1 .67
"1.0
"1.0
~1. 0
"l . 0
1.0
12
PSB1
90
74
61.7
1.38
1.0
1.0
1.0
"1 . 0
"1.0
1 3
PSB 1
90
74
61.7
1.72
"1.0
1.0
"1.0
"l. 0
~1. 0
14
PSB1
90
75
61.7
1.62
"1.0
~1. 0
~1 . 0
~1 . 0
"1.0
15
PSB 1
90
74
61.7
1 . 67
6 . 5
1.0
10.5
8.0
1.0
16
PSfll
90
75
78.1
2.30
"1.0
1.0
~1 . 0
~1. 0
"1.0
1 7
PSB1
90
75
78.1
2.19
"1.0
"1.0
"l. 0
"l. 0
~1 . 0
18
PSB 1
90
73
78.1
2 .24
"l. 0
"1.0
~1. 0
"l . 0
"1 . 0
1 9
PSfll
91
73
78.0
2 .47
"l. 0
"1.0
~1. 0
"l. 0
"1 . 0
20
PSfll
92
74
78 .0
2 . 17
"l. 0
"1.0
~1. 0
"l . 0
"1 . 0
21
PSfll
92
74
78 . 0
1 . 86
"1.0
"1.0
~1. 0
"1 . 0
~1 .0
22
PSB1
93
72
77 .9
2.00
1.0
1.0
1.0
"l. 0
1 . 0
23
PSB 1
93
72
77.9
2.10
"l. 0
1.0
~1. 0
~1. 0
"1 . 0
24
PSfll
93
74
77.9
2.17
"1.0
1 . 0
~1 . 0
"1.0
"l . 0
25
PSB 1
93
75
77.9
2.24
7.4
"1.0
9 . 5
9.0
~1 . 0
26
PSfll
90
72
103.3
3.71
1.0
1.0
"l. 0
1 . 0
1 . 0
27
PSfll
93
73
103.1
3.09
"1.0
"l . 0
"l. 0
"l. 0
"l . 0
28
PSfll
96
73
102 . 8
2.78
"l . 0
"1 . 0
"l . 0
"1.0
~1 . 0
29
PSfll
97
73
102.7
2.72
"l. 0
"1.0
"l . 0
"l. 0
"1.0
30
PSfll
97
74
102.7
2.78
1.9
"l. 0
17.6
17.0
"1 . 0
1KST_TYPE
P-PROPANE
K-KEROSENE
E-EMISSIONS
S-STABILIT1
B-BURNER
L-LOOP
1-STAND	1
2-STAND	2
NL-NORMAL LEAN
NR-NORMAL RICH
LB-LEAN BURNING
LL-LEAN LIMIT
LO-LEAN GOES OUT
RB-RICH BURNING
RL-RICH LIMIT
RO-RICH GOES OUT
QQQ.ES
Z£QCEQUML_COMMENTS
PS-STABILITY PRO C ED. CK.
PN-NOX EMISS. PROCT.D. CK.
PH-HOT SAMPLE LINE CK.
FV-VAPOR GENER. OPER. CK.
BURNER
PAGE 1
NOB	CO	HC	FAN	QQMttMXS
PPM PPM	PPM	RUN~PRO DAT BUR
1 . 0
1.0
"1. 0
0.0299
LO
PS
B
0
1 . 0
"1.0
"l .0
0.0340
LO
PS

0
1 . 0
"1.0
"l . 0
0.0309
LO
PS

0
1 . 0
1.0
1. 0
0.0290
LO
PS

0
1. 0
~1. 0
~1. 0
0. 0298
LO
PS

0
1 . 0
"1.0
"l. 0
0.0288
LO
PS

0
1 . 0
~1 . 0
"1 . 0
0.0317
LO
PS
B
0
1 . 0
"1.0
1. 0
0.0405
LB


0
1.0
~1. 0
"1.0
0.0325
LL


0
1 . 0
"1.0
~1 . 0
0.0301
LO


0
1 . 0
~1. 0
~1 . 0
0.0335
LL


0
1 . 0
"l .0
"1.0
0.0277
LO


0
1 . 0
~1 . 0
~1. 0
0.0345
LO


0
1 . 0
"l . 0
~1 .0
0.0325
LL


0
1 . 0
1 . 0
"l . 0
0.0326
LB


0
1 . 0
"l. 0
"l . 0
0.0364
LB


0
1 . 0
~1 . 0
"1.0
0.0347
LL


c
1. 0
"l . 0
~1 . 0
0.0355
LB


0
1.0
~1 . 0
"1.0
0.0391
LB


0
1.0
"l . 0
"1.0
0.0344
LB


0
1. 0
~1 . 0
~1 . 0
0.0295
LO


0
1.0
1 . 0
"l. 0
0.0317
LO


0
1.0
"l . 0
"1.0
0.0333
LL


0
1.0
"l . 0
~1 .0
0.0344
LL


0
1. 0
"l . 0
"l .0
0.0360
LL


0
1. 0
1 . 0
1 . 0
0.0444
LB


0
1.0
"l. 0
"l . 0
0.0371
LB


0
1 . 0
1 . 0
"1.0
0.0334
LO


0
1.0
1 . 0
1 . 0
0.0328
LO


0
1.0
"1.0
~1 . 0
0.0335
LL


0
Mta_comments	burner_cqmments
B-BAILEY N.G.	O-OLD INJECT. CONFIG.
H-HC N.G.	N-NEJ'' INJECT. CONFIG.
N-NOX N.G.
V-VOLUMETRIC N.G.
T-TRANSIENT OPEP.
Table
VI-1

-------
EXPERIMENTAL DATA FROM THE
WN
TEST
TA
TF
WA
WF (
C02V i
C02C
02V
02B
NOT
W.
TYPE
'F
o F .
IB/RR LB/HR
PCT
PCT
PCT
PCT
PPM
31
PSB1
102
75
120.9
3.71
"1.0
"1.0
0
1
"1.0
"1.0
32
PSB1
105
75
120.6
3.41
~1. 0
"1.0
1
•
O
"1.0
1.0
33
PSB 1
108
77
120 .2
3.21
~1. 0
"1.0
"1 . 0
"l. 0
~1. 0
34
PSB 1
95
75
121.6
3.21
"l. 0
~1 . 0
1.0
"l. 0
"l. 0
35
PSB 1
102
73
120.9
3 . 33
6 . 7
"1.0
10.4
9.6
"l. 0
36
PSB 1
112
74
153.8
4.68
"1.0
"1.0
1 . 0
1.0
1.0
37
PSB 1
117
74
153.2
4.45
"1.0
1.0
1 . 0
~1. 0
"l. 0
38
P5S1
120
74
152.8
4 . 31
~1. 0
~1. 0
"1.0
~1. 0
"1.0
39
PSB 1
120
75
152.8
4. 19
~1. 0
"l. 0
1. 0
1.0
1.0
40
PSB 1
120
75
152.8
4.06
"1.0
"1.0
~1. 0
~1 . 0
1.0
41
PSB 1
124
75
152.3
4 .31
6.5
"1.0
10. 5
9 . 6
~1. 0
42
PSB 1
91
75
103.2
3.71
10. 1
9.4
5 . 5
4.9
1.0
43
PSB 1
93
75
103.1
2.78
6.9
"1.0
10.6
10 .0
"l. 0
44
PSB 1
87
73
41 . 8
1.63
1.0
"1.0
"1.0
1.0
1.0
45
PSB 1
85
75
41.9
1.50
"1.0
"1.0
1.0
"1.0
~1 . 0
46
PSB 1
84
75
41 . 9
1.33
"1.0
"1.0
"1.0
~1. 0
1.0
47
PSB 1
82
75
42.0
1.16
"1.0
"1.0
~1. 0
~1 . 0
"1.0
48
PSB 1
82
74
42.1
1.07
"1. 0
6.9
"l. 0
"l. 0
1. 0
49
PSB 1
80
72
42.1
0.98
1.0
5. 8
1 . 0
"1.0
1.0
50
PSS1
80
73
42.1
1.03
6.5
6 . 6
10 .9
10 . 5
~1. 0
51
PSBl
83
73
25.4
0.80
"l. 0
"1.0
1. 0
"1.0
1.0
52
PS31
82
73
25.4
0.71
"1. 0
~1. 0
1 . 0
~1. 0
1. 0
53
PSBl
82
73
25.4
0.61
"1.0
"1.0
1. 0
1. 0
1.0
54
PSB 1
83
73
25.4
0.66
5.9
"1.0
12 . 0
10 . 9
"1.0
5 5
PSBl
83
73
42.0
6.90
~1. 0
~1. 0
"l . 0
~1. 0
1.0
56
PSBl
81
73
42.1
5.75
"1.0
~1. 0
1. 0
1 . 0
1.0
57
PSBl
82
73
42.0
6 . 30
0. 8
0.4
17.3
1.0
1.0
58
PSBl
85
75
41.9
5.75
0. 6
0.7
19 . 3
"1. 0
"1.0
59
PSBl
85
75
41 .9
6.90
0.1
0 . 5
20 . 3
"1.0
~1 . 0
60
PSBl
90
75
41.7
4.20
0.0
0.3
20. 9
"1. 0
"1.0
QQ2ES
T£ST_TYEE
B.M-Q.Q.MRRTS
E&0Q&2UML-G.QMMENTS
P-PROPANE
K-KEROSENE
E-EMISSIONS
S-STABILITY
B-BURNER
L-LOOP
1-STAl/D	1
2-STAND	2
NL-NORMAL LEAN
NR-NORMAL RICH
LB-LEAN BURNING
LL-LEAN LIMIT
LO-LEAN GOES OUT
RB-RICH BURNING
RL-RICH LIMIT
RO-RICH GOES OUT
PS-STABILITY PROCED. CK.
PN-NOX EMISS. PROCED. CK.
FH-HOT SAMPLE LINE CK.
FV-VAPOR GENER. OPER. CK.
BURNER
PAGE 2
NOB	CO	HC
PPM PPM	PPM
1.0
1.0
1
1 . 0
~1 . 0
~1
1.0
~1. 0
~1
1. 0
"1.0
~1
1.0
~1 . 0
"l
1.0
"1.0
"1
1.0
~1 . 0
"l
1. 0
~1 . 0
"l
1.0
1.0
1
1.0
"1.0
~1
1.0
"1. 0
~1
1.0
124 . 0
~1
1.0
~1 . 0
~1
1.0
"1.0
1
1.0
"1. 0
"l
1 . 0
"1.0
~1
1.0
"l. 0
"l
1. 0
8.0
"1
1.0
688 .0
~1
1 . 0
350 .0
~1
1 . 0
~1. 0
1
1. 0
1.0
1
1. 0
"l . 0
1
1.0
"1.0
"1
1. 0
1.0
~1
1 . 0
1. 0
"1
1.0
713.0
1
1 . 0
12200.0
~1
1.0
5530 .0
~1
1. 0
5560.0
"1
MZ£-COtf VENTS
B-BAILEY N.G.
H-HC N.G.
N-NOX N.G.
V-VOLUMETRIC N.G.
T-TRANSIENT OPER.
FAN	QQHHKS.TS
RUN PRO DAT BUR
0.0380
LB

0
0.0350
LB

0
0.0330
LL

0
0.0326
LO

0
0.0332
LL

0
0.0376
LB

0
0.0359
LB

0
0.0349
LB

0
0.0339
LO

0
0.0329
LO

0
0 . 0350
LL

0
0.0484
LB

0
0.0334
LL

0
0.0482
LB

0
0.0443
LB

0
0 .0392
LB

0
0.0341
LB

0
0.0314
LB

0
0.0288
LO

0
0.0320
LL

0
0.0390
LB

0
0.0346
LB

0
0.0297
LL

0
0.0289
LL

0
0.2032
RL

0
0.1690
RB

0
0.1854
RO
V
0
0.1697
RO
V
0
0.2036
RO
V
0
0.1245
RB
V
0
BUMER_QOMMENZS
-OLD INJECT. CONFIG.
-NEl' INJECT. COliriG.
Table
Vl-2
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
N

-------
EXPERIMENTAL DATA FROM THE
RUN
TEST
TA
TF
WA
WF ,
C02V ¦
C02C
02V
02P
NOT
NO.
TYPE
oF
o F ,
LB/HR
LB/HR
PCT
PCT
PCT
PCT
PPM
61
PSB1
SO
75
41.7
4 . 20
5 . 3
5 . 3
0.1
~1. 0
"1.0
62
PSB1
85
72
41. 9
5.75
3.7
3 . 8
0.4
~1 . 0
1. 0
63
PSB 1
87
72
41.6
6 . 90
1. 0
0.5
16.3
"1.0
1. 0
64
PSB 1
85
72
41.9
5.75
5.1
5.3
0 . 1
~1. 0
1.0
6 5
PSB 1
88
75
41 . 8
5 .99
5 . 8
~1. 0
0 . 0
"1.0
"1 . 0
66
PSB 1
85
75
41.9
6 . 20
"1.0
1.0
"1.0
~1. 0
"1. 0
67
PSB 1
85
75
41 . 9
5 .99
5.5
2.6
4.7
~1 . 0
"1.0
68
PSB1
78
75
25 . 5
3.33
0.5
"1.0
17 . 9
~1 . 0
~1 . 0
69
PSB 1
82
75
24 .7
3.07
3.4
4 . 3
6 . 8
~1 . 0
~1 . 0
70
PSB 1
85
78
25.3
3.82
4.5
5.1
5.1
~1 . 0
~1 . 0
71
PSB 1
86
78
25 . 3
4.01
"1.0
"1.0
1.0
"1.0
"1. 0
72
PSB 1
90
78
51.7
7.62
4.7
4 . 3
0 . 0
~1 . 0
1. 0
7 3
PSB 1
93
78
61. 5
7.80
"1.0
"1.0
"1. 0
"1 . 0
"1.0
7 4
PEBl
86
73
25 . 3
0.66
"1.0
6.7
"1.0
~1. 0
"1.0
7 5
PEB1
85
73
25.3
0.95
8.2
9. 0
8 . 3
7.0
~1. 0
76
PEBl
85
75
25.3
1.00
11.2
8.4
3 . 9
4 . 5
"1.0
77
PEB 1
85
75
25.3
1.32
13.3
12.1
0.8
1.0
~1 . 0
78
PEBl
90
75
61.7
1 .78
5 . 3
5.6
12.5
11.1
"1.0
79
PEB 1
92
79
61.6
2.34
9.3
9.2
7.1
5.5
"1.0
80
PEB 1
90
79
61.7
3.11
13.1
12.5
0.9
1 . 0
"1 . 0
81
PEBl
92
80
61. 6
4.60
7.6
7.8
1.4
0 . 2
1 . 0
82
PEBl
103
75
102 .1
2.78
6 . 6
6.9
10 .7
10 . 0
"l. 0
83
PEBl
105
78
101. 9
3.88
10 . 3
2.1
5 . 3
4 . 5
"1.0
84
PEBl
100
72
102. 4
3.95
10.0
10.5
5.6
3.4
"l. 0
85
PEBl
103
77
102.1
4.62
12 . 5
12.1
2 . 0
1 . 9
1 . 0
86
PEBl
105
77
101. 9
6.76
9.2
9. 0
0.0
0.2
"1 . 0
87
PEBl
300
600
59 . 3
1.71
7 .1
6.4
10.1
9.0
1 . 0
88
PEBl
300
600
59.3
2 . 28
9.2
6.8
7.0
4 . 5
1. 0
89
PEB 1
300
600
59 . 3
3.40
10.1
6.2
0.4
0 . 6
1 . 0
90
PEBl
300
720
59.6
1 . 71
7.6
6.5
9 . 3
8 . 0
"1.0
test_tzpe
P-PROPANE
K-KEROSENE
E-EMISSIONS
S-STABILITY
B-BURNER
L-LOOP
1-STAND	1
2-STAND	2
ME-Q.Q.mM'ZS
NL-NORMAL LEAN
NR-NORMAL RICH
LB-LEAN BURNING
LL-LEAN LIMIT
LO- LEAN GOES OUT
RB-RICH BURNING
RL-RICH LIPIT
RO-RICH GOES OUT
QQU.ES
PR0C£2UE£L_£0!£MENTS
PS-STABILITY PROCED. CK.
PN-NOX EMISS. PROCED. CK.
FH-HOT SAMPLE LINE CK.
FV-VAPOR GENER. OPER. CK.
JXVE BURNER	PAGE 3
NOB	CO	HC	FAN	Q.QMMTS
PPM PPM	PUM	RUN PRO DAT BUR
1.0
131000.0
"1.0
0.1245
RB

0
0 .1
126000.0
"1.0
0.1697
RB
N
0
0.1
382 .0
"1. 0
0 .2040
RO
N
0
0.0
101000.0
"1.0
0 .1697
RB
N
0
"1.0
~1 . 0
"1.0
0.1772
RB

0
"1.0
"l .0
"1 . 0
0 .1829
RL

0
0.1
31500.0
"1.0
0.1767
RB
It
0
"1.0
"l. 0
~1. 0
0.1617
RB
V
0
0.1
140000.0
4385.5
0.1535
RB
NH
0
0.1
115000.016971.2
0. 1867
RL
NH
0
"1.0
"1. 0
"1 . 0
0.1961
RO

0
0.0
132000 .0
542 0.0
0. 1527
RO
NH
0
"1.0
"1.0
"1. 0
0.1568
RO

0
0.0
42.0
108.3
0.0323
LL
Nil
0
2.7
8.4
~1. 0
0.0398
NL
N
0
14.5
18 . 8
95 . 8
0.0534
NL
NR
0
17.0
471.0
53.4
0.0637
NL
NH
0
0 . 3
3920.0
719.5
0.0357
LL
NH
0
7 . 3
46. 0
23.3
0 . 0440
NL
NH
0
15.5
1510.0
"1.0
0.0635
NR
N
0
0.0
62 800.0
1.0
0.0924
NR
N
0
0.0
3660.0
87.6
0 . 0323
NL
NH
0
11.0
149.0
60.8
0.0491
NL
NH
0
14.4
186. 0
67.3
0.0478
NL
NH
0
16.2
822 . 0
28.9
0.0587
NL
NH
0
1 .4
73900.0
27 .0
0.0820
NR
NH
0
0.9
5 . 0
18.7
0.0345
NL
NF
0
9.4
58.0
3. 5
0 . 0440
NL
NH
0
1 .7
36800.0
21.6
0.0709
NR
NH
0
5.4
16.3
7 . 4
0.0369
NL
NH
0
MT&-COM&ENTS
B-BAILEY N.G.
H-HC N.G.
ll-NOX N.G.
V-VOLUMETRIC N.G.
T-TRANSIENT OPER.
MM&E-QQMMENTS
O-OLD INJECT. CONFIG.
N-NEW INJECT. CONFIG.
Table
VI-3

-------
EXPERIMENTAL DATA FROM THE
RUN
TEST
TA
TF
WA
WF
CO 2V
C02C
027
02B
NOT
NO.
TYPE
o F
o F
LB/HF
LB/RR
PCT
PCT
PCT
PCT
PPM
91
PEB1
310
720
59.3
2.28
11.0
8 . 9
4 . 1
3.0
"1 . 0
92
PEBi
300
720
59 . 3
3.40
7.7
7 . 5
3.0
0 .1
"1.0
93
KEB 2
250
44 0
165.3
5 . 86
6 . 8
5. 5
11.0
9.5
"1. 0
94
KEB2
250
485
165.3
7 . 24
9.3
7.7
7.9
5 . 8
~1 . 0
95
'KEB 2
250
510
165.3
7.98
11.9
8 . 7
4 . 2
3.2
"1.0
9 5.1
PEBI
95
78
107.0
3. 57
8 .1
"1.0
8 . 2
8.0
"l . 0
96
KEB 2
250
520
132.9
7.98
13 .6
9 . 5
0 . 3
0.6
~1. 0
97
KEB 2
95
77
55.9
1.66
"l. 0
"1. 0
"1.0
1.0
1 . 0
98
KEB 2
95
77
55.9
2.10
"1 . 0
~1 . 0
1.0
"1. 0
1. 0
99
KEB 2
95
77
55.9
2.75
"1.0
"1.0
1. 0
~1 . 0
~1 . 0
100
KEB 2
100
77
55.6
3.38
~1. 0
"1. 0
"1. 0
"1.0
~1. 0
101
PEB 2
102
77
55.5
4 .03
~1 . 0
1 . 0
1. 0
1 . 0
"1 . 0
102
PEB 2
102
77
89 . 0
3.62
"1.0
"1 . 0
~1. 0
"1. 0
~1 . 0
103
PEB 2
too
770
50.0
2 .00
1 . 0
"1. 0
"1. 0
3 . 6
"1 . 0
1 0*4
PSB 2
400
800
50.0
1.30
~1 . 0
"1. 0
"1 . 0
10.0
~1. 0
105
PSB2
84
88
49.9
1.73
11.9
1.0
1 . 9
3.2
1 . 0
106
PSB2
85
95
49 . 9
1.32
4 . 8
3 . 5
13.6
10.5
"1. 0
107
PSB 2
85
89
49.9
1. 06
~1. 0
~1 .0
"1.0
"1. 0
1.0
108
PSB 2
85
92
49.9
1.20
5 . 6
"l . 0
12 .0
12.5
~1 . 0
109
PSB 2
85
98
49.9
1.32
7 . 4
"1.0
9.8
9.5
1. 0
110
PSB 2
400
730
50.0
1 .06
5 .1
3.2
12.8
12.1
1 . 0
111
PSB 2
400
730
50 . 0
1. 06
4.1
2.6
14.0
13.0
1 . 0
112
PSB2
400
720
33.2
0.75
6.0
3 . 3
11.4
11.2
1 . 0
113
PSB 2
400
725
33 . 2
0.66
5 .1
2 . 8
13.5
12 . 2
~1 . 0
114
PSB 2
400
730
33.2
0 . 66
4 . 7
2.9
13.2
12.6
1 . 0
i i s
PSB2
400
720
33. 1
0.59
4 . 9
"1 . 0
13 . 2
13 . 0
1 . 0
116
PSB 2
400
710
82.7
2.27
7 . 1
"1.0
10.3
9.9
~1 . 0
117
PSB 2
400
710
82.8
1.87
5 . 6
~1 . 0
12.4
12.0
"1 . 0
118
PSB 2
400
730
82 . 8
1.73
4.6
2.6
13.5
13.1
1 . 0
119
PSB 2
400
720
126.8
2.94
6.1
~1. 0
11.6
11.5"
~1. 0
QQQ.ES
IK§.t_zype
P-PROPANE
K-KEROSENE
E-EMISSIONS
S-STABILITY
B-BURNER
L-LOOP
1-STAND	1
2-STAND	2
RlLE-QomMZS
NL-N0RMAL LEAN
NR-NORMAL RICH
LB-LEAN BURNING
LL-LEAN LIMIT
LO-LEAN GOES OUT
RB-RICH BURNING
RL-RICH LIMIT
RO-RICH GOES OUT
EEQ£EQURAL_COi*i
-------
EXPERIMENTAL DATA FROM THE
RUN
TEST
TA
TP
VA
WF i
C02V i
C02C
02 V
02B
NOT
NO.
TYPE
o F
oF LB/RR .
LB/RR
PCT
PCT
PCT
PCT
PPM
120
FSB 2
400
475
127.0
2. 94
5 . 1
"l. 0
12.6
12.2
~1. 0
121
PSB2
400
450
126.8
2. 87
5 . 7
~1. 0
11.8
11.6
~1 . 0
122
FEB 2
400
450
126.8
2.75
4 . 0
~1. 0
13.8
1. 0
~1. 0
123
PEB 2
74
70
50.5
2 .20
10.6
8.7
4 . 5
4.1
"1 . 0
124
PEB 2
80
78
49 .5
1 .27
1.9
7 . 3
17.0
8.0
"1. 0
125
PEB 2
85
78
47.6
1 .27
4 . 5
3 . 3
13.0
"1.0
"1.0
126
PEB2
95
78
127 . 3
3.68
6.4
5 . 5
10.4
7.5
"1.0
127
PEB2
95
78
119.6
3.68
~1. 0
5 . 3
"1 . 0
~1 . 0
"1.0
128
PEB2
90
100
67 . 8
1.99
~1. 0
"1.0
1 . 0
1. 8
1 . 0
129
PEB 2
80
80
86 .2
2 . 36
"1.0
"1.0
"1.0
~1. 0
"1.0
130
PEB 2
80
80
82 . 8
2.61
1. 0
~1. 0
1 . 0
"1.0
1. 0
131
PEB 2
80
80
66 . 7
2.48
"1 . 0
"1 . 0
"l .0
1 . 0
"1.0
132
PEB 2
80
80
44.5
2. 54
~1 . 0
~1. 0
1 . 0
1. 2
1.0
133
PEB 2
80
80
83.4
2 . 54
"1.0
"l. 0
~1. 0
6.8
"1.0
134
PEB 2
80
80
66 . 7
2 . 54
"1.0
"1.0
"1.0
2.5
"1 . 0
135
PEB2
80
80
72.3
1 .70
~1. 0
~1 . 0
"1.0
9.5
~1. 0
136
PEB 2
88
90
72.3
2 . 08
"1.0
"1. 0
"1.0
7 . 5
"l . 0
137
PEB 2
90
90
66 . 7
2 . 54
~1 . 0
"l. 0
"1.0
2 . 6
"l . 0
138
PEB 2
95
95
96 . 6
2.95
8 . 5
"1.0
7.9
7.6
"1 . 0
139
PEB 2
95
100
95.0
3.73
"1. 0
"l . 0
"1.0
3 . 8
"1. 0
140
PEB 2
95
100
91.1
5.06
"l. 0
"1. 0
"1 . 0
1 . 0
"1. 0
141
PEB 2
95
100
77 . 3
5.06
"l. 0
"l. 0
"1.0
"l . 0
~1 . 0
142
PEB 2
95
100
77 . 8
2 .27
~1. 0
"l. 0
"1.0
1 . 0
~1 . 0
143
PEB 2
95
100
77 . 8
5.06
"l. 0
"1 . 0
"1 . 0
~1. 0
~1 . 0
144
PEB 2
95
100
91 .1
5.06
1. 0
"l. 0
~1 . 0
"1.0
"l . 0
145
PEB 2
95
100
94.4
3.73
~1 . 0
"1 . 0
"1.0
~1. 0
~1 . 0
146
PEB 2
95
100
96 . 6
2 . 84
"1.0
~1. 0
"l . 0
9.5
"l . 0
147
PEB 2
95
100
98 . 3
2. 50
"1. 0
"l. 0
"1.0
10 . 8
"1 . 0
148
PEB 2
95
100
90 . 5
3.73
~1. 0
"1. 0
~1. 0
5 . 0
"l . 0
149
PEB 2
95
100
53 . 6
3.73
7 . 4
"1. 0
"1. 0
0.0
"1.0
test_type
P-PROPANE
K-KEROSENE
E-EMISSIONS
S-STABILITY
B-BURNER
L-LOOP
1-STAND	1
2-STAND	2
&M-COWENTS
NL-NORMAL LEAN
NR-NORMAL RICH
LB-LEAN BURNING
LL-LEAN LIMIT
LO-LEAN GOES OUT
RB-RICH BURNING
RL-RICH LIMIT
RO-RICll GOES OUT
CODES
RE.QlQ.equkal_coements
PS-STABILITY PROCFD. CM.
PN-NOX EMISS. PR0CED. CK.
FH-HOT SAf'PI.E LINE CK.
FV-VAPOR GENER. OPER. CK.
FC-FACILITY CK.
BURNER
PAGE 5
NOB	CO	nc
PPM	PPM	PPM
"1.0	"1.0	"l.
"1.0	"1.0	~1.
"1.0	"1.0	"l.
14.6	18.0	"l.
1.5	<15 0.0	~1.
1.2	585.0	"l.
5.2	430.0	"1.
_5.0	630.0	"l.
1.0	"1.0	"l.
"1.0	"1.0	"l.
1.0	1.0	"1 .
"l . 0	"l.0	~1 .
1.0	"1.0	1.
"1.0	"1.0	"1.
"1.0	"1.0	~1.
"1.0	"1.0	369.
"1.0	"1.0	1137.
*1.0	"1.0	14.
9.8	"l.0	"1.
~1 .0	~1.0	"1 .
"1.0	"1.0	"l.
"1.0	"1.0	"l.
"1.0	"1.0	"1.
"1.0	"l.0	~1.
"1.0	"1.0	"l.
"1.0	"1.0	"l.
"l . 0	"l.0	"l.
"1.0	"1.0	"1.
~1.0	"l.0	"l.
0.0	"1.0	"1.
DAlA_COMilENZS
3-BAILEY N.G.
H-HC N.G.
N-NOX N.G.
V-VOLUMETRIC N.G.
T-TRANSIENT OPER.
FAN	QQUMENTS
RUN PRO DAT BUR
0.0274
LB


0
0.0286
LB


0
0.0257
LO


0
0.0509
NR
FC
N
0
0.0304
LL

V
0
0.0316
NL


0
0.0323
LO


0
0.0364
LL


0
0.0346
LB
FH
T
0
0.0324
LB

T
0
0.0373
LB

T
0
0.0440
LB

T
0
0.0675
LB

T
0
0.0360
LB

T
0
0.0450
LB

T
0
0.0278
LB

TH
0
0.0340
LB

TH
0
0.0450
LB

TH
0
0.0410
NL


0
0.0465
NL


0
0.0657
NL


0
0.0774
NR


0
0.0345
NL


0
0.0769
NL


0
0.0657
NL


0
0.0467
NL


0
0.0348
NL


0
0.0301
LL


0
0 . 0487
NL


0
0.0824
NR


0
BURNER-COMMENTS
-OLD INJECT. CONFIC.
-NEV INJECT. CONFIC.
Table
VI-5
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
5
9
7
0
0
0
0
0
0
0
0
0
0
0
0
0
N

-------
EXPERIMENTAL DATA FROM THE
RUN
TEST
TA
TF
WA
WF i
C02V i
C02C
02V
02B
NOT
NO.
TYPE
"F
oF :
LB/HR LB/RR
PCT
PCT
PCT
PCT
PPM
150
PEB 2
95
100
56.9
1.35
"l. 0
"1. 0
"1. 0
10.8
"l. 0
151
PEB2
400
160
84 . 0
3.21
~1. 0
"1. 0
. 0
5 . 8
"1.0
152
PEB2
400
330
79.5
3.21
"l. 0
~1. 0
"1.0
4 . 7
"1 . 0
153
PEB 2
360
300
111.2
2.83
"1.0
"1.0
"1. 0
11 .5
"1.0
154
PEB2
355
200
55 .6
2.11
~1. 0
"1.0
"l . 0
6.0
"1.0
155
PEB 2
270
275
98.9
2.70
~1. 0
"1.0
~1. 0
10 . 0
~1. 0
156
PEB 2
260
255
103.3
2.40
1.0
"1.0
1.0
12.8
"1.0
157
PEB 2
260
290
100.0
3.17
0.1
"l. 0
"1.0
8 . 8
"1 . 0
158
PEB2
260
317
100 . 6
3.47
~1. 0
~1. 0
"1.0
7.5
"1.0
159
PEB 2
265
350
97 . 3
4.00
1.0
"1.0
1 . 0
3 . 8
1.0
160
PEB2
270
410
90 .7
6.00
1. 0
"l. 0
"1. 0
0 . 0
"1. 0
161
PEB 2
275
392
89 .6
6.02
"1.0
~1. 0
~1. 0
0 . 0
~1. 0
162
PEB 2
275
480
87 .9
7.42
~1. 0
"1.0
"1.0
0 . 0
"1.0
163
PEB 2
360
110
22.0
0.65
~1. 0
"1.0
~1. 0
8 . 5
"1.0
164
PEB 2
370
120
22 . 0
0.90
1.0
1.0
1.0
1 . 6
"1.0
165
PEB 2
375
120
30 . 8
1.32
1.0
"1.0
"1. 0
5 . 3
~1 . 0
166
PEB 2
365
140
44 . 5
1.60
"1.0
~1. 0
~1 . 0
7 . 7
~1. 0
167
PEB 2
370
150
44 . 0
2 . 00
~1 . 0
"1.0
"1.0
5 . 0
"1. 0
168
PEB 2
370
170
41.2
2.17
"1.0
~1. 0
~1 . 0
2 . 2
"1.0
169
PEB 2
34 5
180
55 . 0
2.17
"1.0
"1.0
"1.0
6.7
"1.0
170
PEB 2
325
190
69 . 8
2.18
"1. 0
"1.0
~1. 0
9.6
~1. 0
171
PEB 2
305
195
78.1
2.18
1.0
"1.0
"l. 0
10 . 8
"1. 0
172
PEB 2
275
195
97 . 8
2.18
"1. 0
1. 0
1.0
13.0
"1.0
174
PEB 2
400
250
96 .6
3.40
"1.0
~1 . 0
~1. 0
8.2
~1 . 0
175
PEB 2
405
275
93 . 3
4.00
"1.0
"1.0
~1 . 0
4 . 8
~1. 0
176
PEB 2
405
280
93.1
4.40
"1.0
"1. 0
~1 . 0
3.5
"1.0
177
PEB 2
405
200
98.6
2.40
5.4
~1. 0
12.5
12 . 2
~1. 0
178
PEB 2
405
195
95.9
2.81
"1.0
"l . 0
"1. 0
10 . 1
~1 . 0
179
PEB2
350
300
133.7
6. 07
"l. n
"1.0
"1 . 0
3.5
~1. 0
180
PEB 2
345
310
134.6
5.20
"l . 0
"1 . 0
"l . 0
6 . 3
"1.0
QODES
TRST_TY£E
RUE-COUMENZS
PRQCMUEAL-QQMdMTS
P-PROPANE
K-KEROSENE
E-EMISSIONS
S-STABILITY
B-BURNER
L-LOOP
1-STAND	1
2-STAND	2
NL-NORMAL LEAN
NR-NORMAL RICH
LB-LEAN BURNING
LL-LEAN LIMIT
LO-LEAN GOES OUT
RB-RICH BURNING
RL-RICR LIMIT
RO-RICR GOES OUT
PS-STAP I LIT! PROCED. CK.
PN-NOX EMISS. PROCED. CK.
FR-HOT SAMPLE LINE CK.
FV-VAPOR GEUER. OPER. CK.
BURNER
PAGE 6
NOB	CO
PPM PPM
1 . 0
1.0
1.0
"1.0
1.0
"l. 0
1.0
"l. 0
1.0
1.0
1 . 0
~1 . 0
1 . 0
"1.0
1 . 0
"1.0
1.0
"1.0
1. 0
"1.0
1.0
"1.0
1. 0
~1 . 0
1 . 0
1 . 0
1 . 0
"l . 0
1.0
"1.0
1 . 0
"l . 0
1. 0
"1.0
1. 0
"1.0
1.0
"l . 0
1 . 0
~1. 0
1.0
"1.0
1. 0
"1.0
1. 0
"l. 0
1. 0
~1 . 0
1.0
"l. 0
1.0
"1.0
3.6
~1 . 0
1.0
~1 . 0
1.0
"l .0
1 . 0
"l . 0
HC	FAN
PPM
"1.0 0.0281
"1.0 0.0452
1.9 0.0478
_0.0 0.0301
"1.0 0.0449
1.3 0.0323
"1.0 0.0275
0.1 0.0375
1.0	0.0408
1.1	0.0486
"1.0 0.0782
2.1 0.0795
200.0	0.0998
776.1	0.0350
496.2	0.0484
1.3 0.0507
0.5 0.0425
0	.6 0 .0538
1	. 5 0. 0623
0.4 0.0467
0 .1 0 . 0369
0.0 0.0330
9.7 0.0264
0.4 0.0416
0.5 0.0507
0.6 0.0559
0.0 0.0271
0.0 0.0346
0.2 0.0537
0.2 0.0457
COUttENTS
RUN PRO DAT BUR
NL

0
NL
T
0
NL

0
LL

0
NL

0
NL

0
LO

0
LL

0
NL

0
NL

0
NR

0
NR

0
NR

0
LL

0
NL
B
0
NR

0
NL

0
NR

0
NR

0
NL

0
NL

0
NL

0
LO

0
NL

0
NL

0
NR

0
NL

0
NL

0
Nit

0
NL

0
DATA _CO(il£E NTS
B-BAILE? N.C..
f'-HC N.G.
N-NOX N.G.
V-VOLUMETRIC N.G.
"-TRANSIENT OPER.
BURNER_COMMENTS
O-OLD INJECT. CONFIG.
ti-KEV INJECT. CONFIG.
Table
VI-6

-------
EXPERIMENTAL DATA FROM THE
RUN
TEST
TA
TF
WA
WF
C02V
C02C
02 V
02P
NOT
NO.
TYPE
• F
o F .
LB/HR .
LB/HR
PCT
PCT
PCT
PCT
PPM
181
PEB 2
340
305
136.1
4.57
"l . 0
~1. 0
"l. 0
8.2
"l. 0
182
PEB 2
340
303
138.8
4.10
"1 . 0
~1. 0
"1 . 0
9 .6
1 . 0
183
PEB 2
335
295
139.0
3.68
~1 . 0
~1. 0
"l. 0
11.0
"1.0
184
PEB 2
334
280
141.7
3.40
"l. 0
~1. 0
"l. 0
11.8
"1. 0
185
PEB 2
335
320
139. 0
3.77
1
O
~1. 0
"1 . 0
10 .7
"1. 0
186
PEB 2
335
340
139.0
4.22
7 . 4
~1 . 0
9 . 5
9.1
"l. 0
188
KSB2
430
340
77 . 9
3 . 30
9 . 6
~1. 0
6 . 7
5.0
~1 . 0
189
KSB2
430
320
79 . 0
2. 77
7.7
~1. 0
9.4
9.0
~1. 0
190
KSB 2
430
300
79 . 0
2.35
"l . 0
~1. 0
"l. 0
11 . 2
~1 . 0
191
KSB2
400
295
66 . 9
2.35
~1 . 0
~1. 0
"1 . 0
9 . 5
~1. 0
192
KSB2
405
295
56.0
2.35
"l . 0
~1. 0
"1 . 0
7 . 1
"1.0
193
KSB2
410
360
51.6
3.95
"l. 0
~1. 0
"l. 0
~1 . 0
~1. 0
194
KSB 2
410
420
113.0
3.77
7 . 6
~1. 0
9.7
8.5
"1. 0
195
KSB 2
410
410
113.6
3 . 36
~1 . 0
~1 . 0
"1 . 0
9 . 5
"l. 0
196
KSB 2
410
390
115.2
3. 07
1.0
"1.0
1 . 0
11 . 0
1.0
197
KSB 2
430
340
101.4
3 . 07
7 . 3
~1. 0
9 . 9
9.8
~1 . 0
198
KSB2
360
260
41.2
1.65
"1 . 0
~1. 0
"l . 0
8 . 5
~1 . 0
199
KSB2
340
320
38 . 5
2 . 30
0
H
1
~1. 0
"1 . 0
1 .1
"1. 0
200
KSB 2
410
270
38 . 5
1.55
6 . 4
~1 . 0
9 . 2
8.2
"1 . 0
201
PEB 2
86
82
83.2
3.70
~1 . 0
~1 . 0
~1. 0
4 . 0
1 . 0
202
PEB 2
82
82
102.9
4 . 00
"1 . 0
~1 . 0
"1. 0
7 .7
"1 . 0
203
KEB2
110
100
148.0
5.60
"l . 0
6 . 7
"1.0
10.5
"1 . 0
205
PEL 2
313
77
76 . 8
2.90
"1 . 0
~1. 0
"1.0
6.0
"1.0
206
PEL 2
253
77
167.8
5 . 90
"1.0
~1 . 0
"1. 0
7 . 0
"1 . 0
207
PEL 2
245
77
181.5
5 . 50
"l. 0
~1 . 0
1. 0
10.0
3.6
21 7
PEB 2
80
~1
83 . 8
2.74
"l . 0
~1 . 0
"1. 0
11.0
"1 . 0
218
PEB 1
100
70
91.5
2 . 30
7 . 0
6. 1
9. 8
9 . 5
"1 . 0
219
PEB 1
110
70
95.3
3 . 55
10.7
8 .1
5 . 3
4 . 0
~1 . 0
220
PEL 2
100
100
86 . 2
2.73
"l. 0
~1. 0
"1. 0
8 . 0
6 .0
221
PEL2
100
100
116.8
3.64
"1. 0
~1. 0
"l . 0
9.0
~1 . 0
TEST_TYPg
P-PROPANE
K- KEROSENE
E - EM IS SI ON S
S-STABILITY
B-BURNER
L-LOOP
1-STAND	1
2-STAND	2
KUH-QOZIMMZS
NL-NORMAL LEAN
NR-NORMAL RICH
LB-LEAN BURNING
LL-LEAN LIMIT
LO-LEAN GOES OUT
RB-RICH BURNING
RL-RICH LIHIT
RO-RICH GOES OUT
QQ.RES
p RQ.CKQ..MA L _COMMENTS
PS-STABILITY PROCED. CK.
PN-NOX FMIFS. PROCED. CK,
F1I-HOT SAMPLE LINE CK.
PV-VAPOR GENEP. OPER. CK .
LK-LIQUID KEROSENE
BURNER
PAGE 7
NOB	CO	HC
PPM PPM	PPM
~1. 0	~1 . 0	0.
1.0	.0	0.
"l.0	"l. 0	0 .
2.0	~1. 0	0.
"1.0	~1.0	0.
_9.8	"1.0	0.
1.0	1.0	0.
~1.0	"l.0	0 .
"l.0	~1 . 0	0 .
"l. 0	"1.0	0 .
"1.0	"1.0	0.
"1.0	"1.0	670.
"l. 0	~1.0	0.
~1.0	"l.0	0 .
"1.0	"1.0	101.
"l.0	"l.0	11.
"l.0	"l.0	0 .
"l.0	~1.0	0 .
"1.0	"1.0	7.
1.0	1.0	"1.
"1.0	"1.0	~1.
9.8	81.0	0.
"1.0	"1.0	"1.
"l.0	I1-0	I1 •
6.1	1.0	1 .
"1.0	"1.0	~1.
1.9	7.0	~1.
32.6	307.0	"l.
_7.1	"l.0	4.
1.0	1.0	4 .
MZA-Q.Q.MMENTS
B-BAILEY N.G.
H -HC N.G.
N-NOX N.G.
V-VOLUMETRIC N.G.
T-TRANSIFNT OPER.
FAN	COMMENTS
RUN PRO DAT BUR
0.0397
NL


0
0.0349
NL


0
0.0313
NL


0
0.0284
LL


0
0.0321
NL


0
0.0360
NL


0
0.0455
NL


0
0.0369
NL


0
0.0326
LL


0
0.0384
NL


0
0.0460
NL


0
0.0839
NR


0
0.0364
NL


0
0.0324
NL


0
0.0328
LC


0
0.0354
NL


0
0.0439
NL

F
0
0.0655
NR

F
0
0.0441
NL

VF
0
0.0526*
NR


0
0.0460'
NL


0
0.0414
NL
LK

0
0.0447
NL
FV

0
0.0416
NL
FV

0
0.0358
NL
FV

0
0.0387
NL
FV

0
0.0396
NL
PN

0
0.0502
NL
PN

0
0.0375
NL


0
0.0369
NL


0
MMRB.-Q.QilllE.lLTS
-OLD INJECT. CONFIG.
-NEW INJECT. CONFIG.
Table
Vl-7
2
1
1
1
3
0
1
0
0
2
1
0
4
4
5
3
1
9
9
0
0
4
0
0
0
0
0
0
5
6
0
N

-------
EXPERIMENTAL DATA FROM THE
RUN
TEST
TA
TF
UA
WF CO 2 V
C02C
02V
02 P
NOT
NO.
TYPE
o F
»P
LB/HR .
LB/HR
PCT
PCT
PCT
PCT
PPM
22 2
PEL 2
100
100
131.8
3.64
"1 . 0
"1.0
0.1
10 . 0
1. 0
223
PEL 2
100
100
108.4
3.64
~1 . 0
"l. 0
~0 .1
8. 0
"1.0
224
PEL 2
100
100
168.5
5.07
"1.0
~1. 0
"0 . 1
9.0
~1. 0
225
PEL2
100
100
151.8
5.07
"l . 0
1.0
"0.1
8.2
"1. 0
226
PEL 2
100
100
177 . 9
5.07
"l . 0
"1.0
"0 . 1
10.0
~1. 0
227
PEL2
76
82
55.8
1.75
"1. 0
8 . 5
"0 .1
7.5
~1. 0
228
PEL 2
82
90
58.4
1. 82
"l . 0
7.4
"0 .1
11.0
"1.0
229
PEL2
85
95
57.6
1 .80
"l . 0
8.5
~0 .1
9.0
~1. 0
230
PEL2
433
98
57. 6
1. 80
"1.0
~1. 0
"0 . 1
9.5
"1.0
231
PEL 2
420
100
60. 8
1 . 80
~1 . 0
~1. 0
"0 .1
11 . 2
~1. 0
232
PEL 2
410
100
90. 9
1.70
"1.0
~1. 0
"0.1
10.5
"1. 0
233
PEL 2
407
100
90. 2
1.25
~1 .0
1.0
"0 . 1
8. 3
"1.0
231
PEL2
90
82
76.9
2.80
"1.0
"1.0
"0 . 1
4.0
1.0
235
PEL2
92
89
77.3
2.48
~1. 0
"1.0
~0 .1
6.2
~1. 0
236
PEL2
89
89
77 . 7
2.05
7 . 0
~1. 0
10 . 6
8 . 8
~1 . 0
237
PEL2
82
82
38.4
1 . 32
~1 .0
~1. 0
"0 .1
4 .0
53 . 8
238
PEB1
79
70
49.1
1.55
5.4
"l. 0
12 .4
10 . 0
"1.0
239
PEB1
80
71
48.9
1.90
7.4
1.0
10.2
8. 0
1.0
210
PEBl
82
71
48.9
2 .04
8.7
~1. 0
7.5
6 . 0
"1.0
241
PEB 1
82
72
48. 8
2 . 50
9.7
~1. 0
6 .0
4 . 0
"1.0
242
PEB 1
78
68
49 . 2
2.45
9.9
~1 . 0
4 . 8
4.0
"1.0
243
PEB 1
80
72
48 . 8
2 .62
11.7
~1. 0
2.4
2.0
1 . 0
244
PEB 1
79
68
98. 1
3.36
6 . 5
1. 0
10.7
10 .0
"1. 0
245
PEBl
80
70
97.6
3.88
7 . 8
"l. 0
8.6
8 . 0
~1. 0
246
PEB 1
80
70
60 . 5
2.50
7 . 2
1. 0
9.5
12.0
"1.0
247
PEB 1
75
70
"1. 0
"l .00
~1 . 0
~1 . 0
"0 . 1
11.8
1 . 0
248
PEBl
85
70
47 . 2
"1 .00
~1. 0
1. 0
"0 . 1
9 . 5
"1.0
249
PEBl
75
70
49.2
2.70
7 . 3
1 . 0
9.6
9.5
"1.0
250
PEBl
85
70
48 . 8
2.70
o
o
H
"l. 0
"0 .1
5 . 1
~1 . 0
251
PEB 1
85
70
100.8
3.25
5 . 6
1.0
14 . 3
1J . 0
"1. 0
TES£_TYPE
P-PROPANE
K-KEROSENE
E-EMISSIONS
S-STABILITY
B-BURNER
L-LOOP
1-STAND	1
2-STAND	2
NL-NORMAL LEAN
NR-NORMAL RICH
LB-LEAN BURNING
LL-LEAN LIMIT
LO-LEAN GOES OUT
PB-RICH BURNING
RL-RICH LIMIT
RO-RICR GOES OUT
Q.QRRS
ERQ£EQUli&i,_CQilMENTS
PS-STABILITY PROCED. CK.
PK-NOX EMISS. PROCED. CK.
FH-HOT SAMPLE LINF CK.
FV-VAPOP GENER. OPER. CK.
FC-FACILITY CK.
LK-LIOUID KEROSENE RUfr
BURNER
PAGE 8
NOB	CO	HC	FAN	COMygNTS
PPM
PPM
PPM

RUN
PRO DAT BUP
"1. 0
1.0
24.0
0.0327
NL

0
"1.0
~1. 0
7.9
.0.0397
NL

0
"l. 0
~1. 0
4.3
0.0356
NL

0
"l. 0
"1.0
12 . 9
0.0395
NL

0
"1.0
1.0
~1 . 0
0.0337
LL

0
17.2
7.0
12.6
0.0371
NL

0
2.4
7 . 0
5.4
0.0369
NL

F 0
7.6
7.0
8.5
0.0369
NL

F 0
5.0
"1.0
5.9
0.0370
NL

F 0
2.7
~1. 0
4.1
0.0350
NL

F 0
4 .1
1. 0
3.4
0.0333
NL

F 0
13. 8
1 . 0
6.4
0.0395
NL

F 0
36.6
1.0
24 . 0
0.0430
NL

0
22.4
"1. 0
35.5
0.0380
NL

0
15.0
"l. 0
40.4
0.0384
NL

0
42.6
"l . 0
42 . 3
0.0407
NL

0
1.3
"l. 0
"1.0
0.0312
NL

0
4.2
1.0
1. 0
0.0345
NL

0
12.6
"l. 0
1. 0
0.0420
NL

0
60.0
"1.0
1. 0
0.0465
NL

0
58 . 7
~1. 0
~1. 0
0.0488
NL

0
104 . 0
~1. 0
"1 . 0
0.0573
NL

0
2 . 5
1. 0
1. 0
0.0323
NL

0
6 . 6
~1 . 0
1. 0
0.0384
NL

0
"1.0
"1.0
1.0
0.0409
NOT
VALID
DATA
"1.0
~1. 0
1. 0
0.0295
NOT
VALID
DATA
"l . 0
~1. 0
~1. 0
0.0360
NOT
VALID
DATA
4. 1
~1 . 0
*1.0
0.0542
NOT
VALID
DATA
46 .5
~1. 0
1.0
0.0548
NOT
VALID
DATA
1.6
"1 . 0
1.0
0.0319
NOT
VALID
DATA
UATA_COHdENTS
B-BAILEY N.G.
H-HC N.G.
N-NOX N.G.
V-VOLUMETPIC N.G.
F-FLOl' N.G.
T-TRANSIENT OPER.
Table
VI-8
MMer_comments
O-OLD INJECT. CONFIG.
N-NEW INJECT. CONFIG.

-------
EXPERIMENTAL DAT& FROM THE
RUN
TEST
TA
TP
WA
WF i
C02V
C02C
02V
02B
NOT
NO.
TYPE
«F
°F .
LB/HR .
LB/HP
PCT
PCT
PCT
PC?
PPM
252
PEBX
90
70
100.4
3.25
~1. 0
"l . 0
"0.1
9.5
~X . 0
253
PEB1
92
70
97.0
3.62
7.0
1. 0
7.8
8.0
~1. 0
254
PEB1
93
70
100.1******
8 . 5
"1.0
7.8
6.3
"l. 0
255
PEL2
93
70
90.0
2.18
8 . 0
"1.0
7.0
"l. 0
"1.0
256
PEBX
85
70
97.6
1. 00
9 . 5
"1 . 0
6.3
6.0
~1. 0
257
PEB 1
85
70
97 . 6
1 . 00
10.5
"1 . 0
4 . 8
4 . 0
1.0
258
PEB 1
85
70
97.5
5.25
12.4
~1 . 0
2.6
2.0
~1 . 0
259
PEB X
85
70
97.4
"1.00
12 . 9
"1. 0
0.0
0 . 0
"l. 0
260
PEB 1
85
70
97. 3
6.10
12.2
"1. 0
0.5
1.0
"1 . 0
261
PEB 1
89
70
97.2
"l. 00
9.5
"l . 0
0.9
0.2
"1. 0
262
PEB 1
90
70
144.4
4 .83
6.7
"1. 0
10.5
10 . 0
"l. 0
263
PEB 1
97
70
143.5
5.75
8.0
1 . 0
9 .1
8 . 0
1. 0
2 64
PEB 1
85
73
48.8
1.68
6.6
6 .0
11.1
11.3
"l. 0
265
PEB 1
85
73
48.8
~1. 00
7.7
7 . 5
9.2
9.2
"1. 0
266
PEB 1
96
70
143.0
4.82
7.4
9 . 8
9.3
8.9
1.0
267
PEBX
96
70
97.8
3.20
6.9
"l. 0
11.2
11.0
"l. 0
268
PEBX
96
70
97.2
3. 55
7.7
"l. 0
9.2
9 . 0
"1 . 0
269
PEBX
96
72
97.2
4.15
8.7
"1. 0
7.4
7.0
1. 0
270
PEBX
96
72
97.1
4.30
9.0
"1 . 0
6.7
6.0
~1. 0
271
PEBX
100
75
144 . 1
5 .10
7.1
"l. 0
9.4
9.4
~1. 0
272
PEBX
100
71
141.7
5.95
8.9
"1. 0
6.9
6.9
~1. 0
273
PEBX
100
70
143.1
6.65
9 . 0
"1. 0
7.2
6 . 0
~i . 0
274
PEBX
100
70
143.5
6.60
10 .2
9 . 7
5.5
5.4
"1. 0
275
PEBX
95
70
143.7
7 . 30
11.1
10 . 3
4.1
3.8
"1 . 0
279
PEL2
109
70
91.0
2.82
7 . 4
7 . 7
8.8
8.1
4.0
280
PEL2
110
70
90.9
3.35
~X . 0
8 . 3
~1. 0
5.9
20 . 8
281
PEL2
110
70
111.3
3.46
7.7
8.0
8.7
8.3
4.1
282
PEL2
94
70
4 6.4
1. 35
8 . 0
8.5
8.6
8.0
7.7
283
PEL 2
100
70
47.3
1.25
7. 3
6 .0
10.0
9.5
2 . 0
284
PEL2
105
70
46.5
1 . 08
5.4
5 . 8
12.5
11. 3
0 . 8
T£ST_TZZ£
P-PROPANE
K-KEROSENE
E-EMISSIONS
S-STABILITY
B-BURNER
L-LOOP
1-STAND	X
2-STAND	2
RM-QQ.MMENTS
NL-NORMAL LEAN
NR-NORMAL RICH
LB-LEAN BURNINC
LL-LEAN LIMIT
LO-LEAN GOES OUT
RB-RICH BURNING
RL-RICH LIMIT
RO-RICH GOES OUT
QQQ&S
ERQCEDURAL_COMMENTS
PS-STABILITY PROCED. CK.
PN-NOX EMISS. PROCED. CK.
FU-HOT FAMPLE LINE CK.
FV-VAPOR GENER. OPER. CK.
FC-FACILITY CK.
LK-LIPUID KEROSENE RUN
BUPNER
PAGE 9
NOB	CO	HC	FAN	COMMENTS
PPM PPM	PPM	RUN PRO DAT BUR
1.1
"1.0
~1 . 0 ,
,0.0320
NOT
VALID
DATA
3.1
"1.0
~1 . 0 !
0.0369
NOT
VALID
DATA
6.6
"1.0
1. 0
0.0410
NOT
VALID
DATA
"l. 0
~1 . 0
"l . 0
0.0388
NL
FV
0
14 . 7
"1.0
"l . 0
0.0460
NL

0
44. 0
"1.0
"1 . 0
0.0503
NR

0
78.5
"1.0
"1 . 0
0.0576
NR

0
113.0
"1.0
"l . 0
0.0640
NR

0
41.0
1.0
"1.0
0.0615
NR

0
59.0
~1 . 0
~1 . 0
0.0760
NR

0
5 . 0
"1.0
"1 . 0
0.0331
NL

0
9.4
1.0
1.0
0.0381
NL

0
2.2
5 . 0
"1. 0
0.0317
NL

0
10.2
7.0
"1 . 0
0.0372
NL

0
~1. 0
10. 0
1 . 0
0.0363
NL

0
2. 0
"1.0
"l. 0
0.0323
NL

0
6 . 6
~1. 0
"l. 0
0.0372
NL

0
19 . 0
"1.0
"l . 0
0.0421
NL

0
56.5
"1.0
"1.0
0.0439
NL

0
4 . 2
"1.0
~1 . 0
0.0357
NL

N
14 . 1
"1.0
"1.0
0.0424
NL

N
42 . 0
"1.0
1 . 0
0.0432
NL

N
12.5
24 .5
0.0
0.0484
NL

N
3 6.5
42.7
~1 . 0
0.0528
NL

N
8.2
5 . 0
2.6
0.0371
NL

N
22 . 6
10 . 0
14.5
0.0436
NL

N
8 . 3
5. 0
0 . 6
0.0380
NL

N
9 . 4
5 . 0
0 . 0
0.0389
NL

N
2.8
5.0
0 . 0
0. 0386
NL

N
1. 1
5 . 0
0 . 0
0.0271
NL

N
MZ6._comments	burner-Comments
B-BAILEY N.G.	O-OLD INJECT. CONFIG.
H-HC N.G.	N-NEW IKJFCT. CONFIG.
N-NOX N.G.
V-VOLUMETRIC N.G.
F-FLOW N.G.
T-TRANSIEHT OPER.
Table
VI-9

-------
EXPERIMENTAL DATA FROM THE
RUN
TEST
TA
TF
WA
WF C02V l
C02C
02 V
025
NOT
NO.
TYPE
o F
«F ,
LB/HP LB/HR
PCT
PCT
PCT
PCT
PPM
285
PEL 2
116
70
136.5
3.30
6.4
6.3
11.1
11.0
1.2
286
PEB1
82
75
58.8
1.95
6.3
"1.0
10 . 5
9.6
"1.0
287
PEB1
92
70
144.0
4. 90
5 . 5
6.9
11 .6
10.5
~1 .0
288
PEB 1
90
69
144. 3
5 . 07
7.4
7.0
9.4
9.3
"l . 0
289
PEB 1
95
69
143. 6
5.48
8.0
7.5
8 . 9
8 . 5
"1 . 0
290
PEB 1
99
60
142.9
6.30
9.5
9.0
6 . 1
5.6
"l . 0
291
PEB 1
73
68
66 . 0
5.15
"l. 0
~1. 0
~1. 0
1.0
"1.0
292
PEB 1
75
70
49 . 1
4. 00
"l. 0
0.3
~1 . 0
"l. 0
"1.0
293
PEB 1
82
70
47 . 1
3.85
7.0
6.3
0 . 3
"1.0
"1 . 0
294
PEB 1
90
70
51.1
3.45
8.4
9.0
2 . 4
"1.0
"1.0
295
PEB1
72
70
49 . 2
3.85
7 . 5
7.5
1.0
"1.0
"l . 0
296
PEB 1
86
73
48.6
3.65
7.2
8.0
0.2
~1. 0
"1.0
297
PEB 1
92
71
48.3
4 . 30
6 . 5
6.5
0 . 0
1.0
"l. 0
298
PEB 1
95
71
48.2
3.73
8.1
9.0
2 . 2
"1.0
"l . 0
299
PEB 1
96
71
87.2
8.00
"1.0
7.5
"1.0
"1.0
~1 . 0
300
PEB 1
85
69
49 . 3
4.31
1. 0
8.0
"1.0
"1.0
"1 . 0
301
PEB 1
88
69
48.5
3.40
9.0
10 .1
2 . 8
"1.0
"l . 0
302
PEB 1
92
70
96.9
8.75
6.4
6.5
0 . 8
~1. 0
"1.0
303
PEB 1
74
68
98.5
7.30
8.0
7.9
0.0
1.0
"l . 0
304
PEB 1
87
70
141. 5
10.92
7.2
7.4
1 .5
"l. 0
"1. 0
305
PEB 1
100
67
124.2
11.65
6.9
7 . 0
0 . 0
"1.0
"1 . 0
306
PEB 1
100
68
139.8
10.26
9.8
9.9
0 . 1
~1 .0
"1.0
307
PEB 1
104
69
94.3
6.40
9.4
9.9
0 . 0
"1.0
"1. 0
308
PEL 2
90
320
46.6
1. 60
7.2
6.9
9 . 5
8.6
2 . 3
309
PEL 2
98
350
47 . 5
1. 50
6.7
5.9
11 .0
10 . 3
1 . 3
310
PEL 2
94
310
48.4
0. 90
4.0
5 .0
15.2
12.4
0 . 8
311
PEL2
385
350
49.4
1.7 5
7.7
7.4
8.9
8.5
7.6
312
PEL 2
388
348
48.3
1.30
6.8
6 . 3
10.8
10.5
1.6
313
PEL2
395
340
47.2
1.00
5 . 0
5.0
13.1
12.5
0.6
314
PEL2
340
450
88.2
2.78
7.2
"1.0
10.2
8.6
3.6
test_tzpe
P-PROPANE
K-KEROSENE
E-EMISSIONS
S-STABILITY
B-BURNER
L-LOOP
1-STAND	1
2-STAND	2
EM-COMMENTS
NL -NORMAL LEAN
NR-NORMAL RICH
LB-LEAN BURNING
LL-LEAN LIMIT
LO-LEAN GOES OUT
PB-RICH BURNING
RL-RICU LIMIT
RO-RICR COBS OUT
CQDES
P&QCEDUML_£Q£iiE!lTS
PS-STABILITY PROCED. CK.
PN-NOX EMISS. PROCED. CK.
FH-HOT SAMPLE LINE CK.
FV-VAPOR GENER. OPER. CK.
FC-FACILITY CY.
LK-LICUID KEROSENE RUN
BURNER
PAGE 10
NOB	CO	HC
PPM	PPM	PPM
"1.0	8.2	_0.0
2.3	"1.0	1.0
3.5	95.0	0.0
4.4	"1.0	0.0
6.5	"1.0	0.0
4.0	41.0	_5.9
35.0	"1.0	1.0
26.8	12.5	5000.0
"1.0	98000.0	30.3
"1.0	64500.0	76.0
63.2	83500.0	25.9
90.0	82800.0	17.6
55.6	104000.0	17.0
109.6 75000.0	24.2
"1.0 87500.0	26.5
40.0 81500.0	27.0
117.0 52000.0	15.4
17.8 100000.0	"1.0
30.0 36200.0	65.0
22.7	83400.0	28.9
24.0 91600.0	25.0
30.0 45900.0	157.5
18.8	45500.0	1.0
4.8	75.0	1.0
1.9	75.0	0.0
0.8	13.8	1.4
8.4	75.0	0.0
0.4	5.0	0.0
0.1	5.0	0.0
2.0	~1.0	1.0
fan CQMUEKIS
RUN PRO DAT BUR
0.0312
NL

N
0.0321
NL

N
0.0337
NL

N
0.0362
NL

N
0.0385
NL

N
0.0460
NL

N
0.0895
NR

N
0.1000
NR
V
N
0.0870
NR


0.0805
NR

N
0.0845
NR

N
0.0860
NR

N
0.0905
NR

N
0.0820
NP

N
0.0880
NR
V
N
0.0750
NR
V
N
0.0780
NR

N
0.0910
NR

N
0.0820
NR

N
0.0860
NR

N
0.0880
NR

N
0.0750
NR

N
0.0765
NR

N
0.0350
NL

N
0.0323
NL

N
0.0220
LB

N
0.0377
NL

N
0.0326
NL

N
0.0252
NL

N
0.0345
NL

N
data_comments
MMZR-QQMKNTS
B-BAILEV N.G.
H-HC N.G.
N-NOX N.G.
V-VOLUMETRIC N.G.
F-FLOV N.G.
T-TRANSIENT OPER.
O-OLD INJECT.
N-NEW INJECT.
CONFIG.
CONFIG.
Table
VI-lo

-------
EXPERIMENTAL DATA FROM THE
RUN
TEST
TA
TP
WA
WF i
CO 2 V
C02C
02 V
02B
NOT
NO.
TYPE
«F
»F ;
LB/HR LB/UR
PCT
PCT
PCT
PCT
PPM
31 5
PEL 2
364
450
93.1
2.40
6.1
"1.0
11.4
10 .4
1.0
316
PEL 2
400
460
90.5
2.16
6. 5
6 . 1
10 . 7
10 . 3
0.7
317
PEL2
435
460
90.0
1.95
5.0
5 . 3
12.6
12.2
0.4
31 e
PEL2
350
638
137. 1
3.75
6 . 4
6.4
11.0
10 . 3
1.8
319
PEL2
350
620
137.2
3.08
4 .8
5 . 3
12 . 8
12 . 3
0.6
320
PEL 2
351
600
134.5
4.47
7.9
7 . 6
6 .6
8 . 6
3.6
321
PEL2
140
540
89 . 8
2.95
7.7
7.9
8.5
8.4
3 . 5
322
PEL2
115
490
90.0
2 . 34
6.2
6. 0
11.0
10 . 3
0.8
323
PEL 2
105
530
90.0
2.18
5 .2
5.6
12.2
11.9
0.6
32 4
PEL 2
9 5
585
135.5
4.45
7.6
7 . 1
8.9
8. 5
2.8
325
PEL2
105
550
137.3
3.85
5 . 7
6 . 3
11.1
10 . 2
1.0
326
PEL2
110
610
138.7
3.60
5 . 7
5 . 9
12 . 0
11.0
0.8
327
PEL2
41 2
77
48.2
1.50
7 . 8
8.1
8.5
8.4
6.0
328
PEL 2
395
90
47. 3
1.15
6.0
6 . 1
11.6
11.2
1. 3
329
PEL2
390
95
47.1
0 . 87
4 . 8
5.0
13.3
12.6
0.4
330
PEL2
407
97
90 . 3
2.93
7.7
7 . 8
8.6
8. 3
9 . 2
331
PEL2
400
100
92.7
2.30
6 .2
6. 3
11.1
10.5
1 . 8
332
PEL2
435
92
89 .7
2.02
5 . 0
5 . 5
12.6
11.0
7.0
333
PEL2
392
95
135.2
3.22
5 . 8
1 . 2
9 . 2
10 .6
1. 1
334
PEL 2
39 5
100
140 . 3
2 .89
4.8
"1.0
13.2
12.2
0 . 7
335
KEL2
92
590
91.3
2.75
7 . 8
7. 8
9.5
8 . 5
3.1
336
KEL2
90
630
92.6
2.41
6 .1
6 . 3
11 . 2
10 . 3
1.2
337
KEL2
90
610
48. 5
1.85
7 . 1
7.3
9.5
8.2
2.4
338
KEL 2
90
660
49.6
1.78
6.0
~1. 0
9.4
9.2
1. 5
339
KEL2
100
660
136 . 7
4.46
7.5
"1.0
8.9
8.4
3.9
34 0
PEL 2
39 5
90
138. 3
3.79
6 . 5
6.6
10 .9
8 . 6
2 . 3
341
PEL2
150
100
92.4
2 .55
6.9
6 . 6
10 .5
8 . a
2.1
342
KEL 2
90
85
93.3
2 .88
7.2
7 . 1
9 . 5
8.4
19.8
343
PEL2
102
100
91.3
2.54
6.7
7.0
10 .1
10.4
2 . 1
344
PEL2
100
100
91.9
2.25
6 . 0
6 . 6
11.4
11.4
1.4
TESZ_ZIP£
P-PROPANE
K-KEROSENE
E- EMISSIONS
S-STABILITY
B-BURNER
L-LOOP
1-STAND	1
2-STAND	2
RUlL.ZQlttiKUZS.
NL-NORMAL LEAN
NR-NORMAL RICH
LB-LEAN BURNING
LL-LEAN LIMIT
LO-LEAN GOES OUT
RB-RICH BURNING
RL-RICH LIMIT
RO-RICH GOES OUT
GQB.ES
£Bi!l£.£K£
PS-STABILITY PROCED. CK.
PN-NOX EMISS. PROCED. CK.
FH-HOT SAMPLE LINE CK.
FV-VAPOR GENER. OPER. CK.
FC-FACILIT Y CK.
LK-LI QUID KEROSENE RUN
'E BURNER
PAGE 11
NOB	CO	HC	FAN	COMMENTS
PPM PPM	PPM	RUN PRO DAT BUR
1.0
1 . 0
1. 0
0.0302
NL
N
1.0
5 . 0
0 . 0
0.0323
NL
N
0.1
5 . 0
0 . 0
0.0260
NL
N
0.8
5 . 0
0 . 0
0.0316
NL
N
0 . 0
9 .2
0 . 0
0.0266
NL
N
1.9
8 . 3
0 . 0
0.0415
NL
N
2.4
15 .0
0 . 0
0.0383
NL
N
0 . 1
16.7
0 . 0
0.0311
NL
N
1.0
158.5
2 . 8
0.0287
LL
N
1.5
5 . 0
0. 3
0.0374
NL
N
0.4
33.5
0 . 1
0.0332
NL
N
0.1
150.0
2.7
0.0307
LL
N
3 . 7
16 . 8
0 . 4
0.0385
NL
N
0.6
15 . 9
0.0
0.0297
NL
N
0.0
82.4
0 . 0
0.0245
LB
N
6 . 7
5.0
0.0
0.0381
NL
N
1.8
5.0
0 . 0
0.0309
NL
N
0 . 0
9930 .0
0. 0
0.0260
LL
N
0.0
144 . 0
0 . 0
0.0282
NL
N
0.0
~1 . 0
0 . 0
0. 0246
LL
N
1 . 8
5 . 0
0.0
0.0370
NL
N
0.0
300 .0
0 . 7
0.0311
LL
N
3 . 5
5 . 0
0 . 0
0.0362
NL
N
0.2
5 .0
0 . 0
0.0393
NL
N
4.4
"1.0
2 . 6
0.0379
NL
N
0 . 5
5 .0
0 .1
0.0320
NL
N
0.6
5 . 0
0 . 0
0.0336
NL
N
0.0
5 .0
"l. 0
0.0365
NL LK
N
1. 0
5 .0
0 . 0
0.0337
NL
N
0 . 1
5 . 0
6 . 8
0.0300
NL
N
Q.6.TA_CQ^dlNTS
B-BAILEY N.G.
H-hC N.G.
N-NOX N.G.
V-VOLUMETRIC N.G.
F-FLOW N.G.
T-TRANSIENT OPER.
MMM-QQUUEUXS
O-OLD INJECT.
N-NEW INJECT.
CONFIG.
CONFIG.
Table
VI-11

-------
EXPERIMENTAL DATA FROM THE
RUN
TEST
TA
TF
WA
WF i
C02V i
£702 C
02 V
02B
NOT
NO.
TYPE
»F
«F .
LBIRR LB/UR
PCT
PCT
PCT
PCT
PPM
345
PEL 2
102
95
135.3
3.97
7.0
7.3
9 . 8
9 . 8
3 . 9
346
PEL 2
105
95
138. 2
3.65
6 . 3
6 . 6
11.0
11.2
1 . 9
347
PEL 2
105
97
134 .5
4. 80
8.6
8.9
7.5
8.4
11.7
348
KEL2
107
95
90.1
2. 88
8 . 5
7.1
8.9
9.5
14 .4
349
KEL2
100
100
91.9
2. 60
6.8
6.4
10.3
11.0
3 . 1
350
KEL2
101
105
136.4
4.00
7.9
1.9
9.5
10 . 4
8.2
351
KEL2
105
810
92.4
3. 00
8.7
7.2
8 . 3
8.4
4 . 7
352
KEL2
110
815
93.1
2.65
7. 3
6 . 9
10.9
9 .6
0 . 3
353
KEL2
120
790
137.7
4.45
8.7
7.1
8.3
8.4
5 . 7
354
KEL2
430
630
81 .4
2.6 2
7.6
6 . 8
9.9
9.2
3.7
355
KEL 2
438
610
82.0
2.15
5.9
5.5
12.1
11.2
0 . 8
356
KEL 2
450
625
43.8
1.70
7.7
7.0
9.4
8.4
3 . 7
357
KEL2
140
69 0
180.5
4.97
7.0
6 . 5
10 .5
9.7
1.7
358
KEL 2
110
690
46 . 8
1.64
6 .4
6.2
11 . 5
8.3
1 .6
359
KEL2
120
730
138.2
4.18
7.1
6 . 9
10 . 8
8.2
2 . 2
360
KEL 2
380
570
92 .6
3. 00
7.5
6 . 8
"1.0
8 . 7
3 . 8
361
KEL2
400
620
90 .9
2.65
6.5
6.1
11 . 5
10.4
1.6
362
KEL 2
405
670
89 . 8
3.30
9.0
9.2
8.0
7.6
8.9
363
KEL2
405
645
90.8
3.05
6.9
7. 0
10 . 8
10 . 5
5 . 0
364
KEL 2
405
670
48.7
1.90
8 .0
7.0
9.2
10.5
4 . 8
365
KEL 2
405
670
48.0
2.10
8 . 8
7.9
8.1
9 . 5
5 .2
366
KEL2
410
660
47.9
1.55
5.9
5 . 6
12.1
13.5
1 . 2
367
KEL 2
405
670
137.0
4.00
7.0
6. 5
10 . 3
12.5
2 . 8
368
KEL 2
410
690
137.1
3.48
5.9
~1. 0
12. 1
13.6
0 . 8
369
KEL 2
400
690
136.3
4. 36
8 . 0
7.6
9.2
11.5
5 . 6
370
KSB 2
403
710
90.9
2.98
7.7
1. 0
9 . 8
9.5
1 . 0
371
KSB2
426
720
137.1
4.36
8 .1
7.5
9.1
9 . 5
"1 . 0
372
KSB 2
425
700
138.7
4.27
7 . 5
6 . 6
9.5
10 . 0
"1 . 0
373
KSB 2
425
720
90.9
2.72
7.5
6 .6
10 .1
9 . 8
"l. 0
374
KSB 2
430
700
90.3
2.62
6 . 8
6 . 1
10 . 7
10.5
"l . 0
T&ST_TYP£
P-PROPANE
K-KEROSENE
E-EMISSI0I1S
S-STABILITY
B-BURNER
L-LOOP
1-STAND	1
2-STAND	2
RUli-QQtitl£?LZS
NL-NORMAL LEAN
NR-NORMAL RICH
LB-LEAN BURNING
LL-LEAN LIMIT
LO-LEAN GOES OUT
RB-RICH BURNING
RL-RICH LIMIT
RO-RICH GOES OUT
QQ.RE.S
EEQCK2UEAL_C0MHMIS
PS-STABILITY PROCED. CK.
PN-NOX EMISS. PROCED. CK.
FH-HOT SAMPLE LINE CK.
FV-VAPOR GENER. OPER. CK.
FC-FACILITY CK.
LK-LIQUID KEROSENE RUN
BURNER
PAGE 12
NOB	CO
PPM PPM
0.6
5.0
0 . 0
1.0
5 . 8
"l .0
0.7
37.5
0.1
4680.0
1.0
555 .0
2.6
5.0
0.4
5 . 0
4 . 0
5.0
2 . 3
5 .0
0.2
5 .0
2.4
5 .0
0.6
5.0
2.0
5 . 0
1 . 4
5 .0
2.5
5 . 0
0.3
5 . 0
5 . 3
5 . 0
3.7
5 . 0
4.4
5 . 0
7 . 2
5 . 0
1 . 0
5.0
1 . 1
~1 . 0
0 . 5
5 . 0
3.6
10.0
2.4
14.6
3 . 8
37. 3
1. 2
46.2
2.2
5 . 0
1.2
205 .0
EC	FAN
'PM
0.0	0.0348
2.0	0.0313
0.0	0.0417
0.0	0.0401
30.1	0.0342
1.3	0.0380
0.0	0.0421
0.0	0.0336
19.3	0.0421
0.0	0.0359
0.0	0.0287
0.0	0.0371
0.0	0.0337
0.0	0.0312
0.0	0.0335
0.0	0.0355
0.0	0.0312
0.0	0.0420
0.7	0.0332
6.6	0.0379
0.9	0.0416
0.0	0.0287
0.0	0.0340
"1.0	0.0287
0.0	0.0379
1.9	0.0365
0.0	0.0383
6.6	0.0362
0.9	0.0355
74.1	0.0327
COMMENTS
RUN PRO DAT BUR
NL

N
NL

N
NL

N
NL
LK
N
NL
LK
N
NL
LK
N
NL

N
LL

N
NL

N
NL

N
NL

N
NL

N
NL

N
NL

N
NL

N
NL

N
NL

N
NL

N
NL

N
NL

N
NL

N
NL

N
NL

N
NL

N
NL

N
LB

N
LB

N
LL

N
LB

N
LL

N
DATA_Q.OmE.NTS
UUMKR-COMMENTS
B-BAILEY N.C.
H-HC N.C.
N-NOX N.G.
V-VOLUMETRIC N.G.
F-FLOW N.G.
T-TRAUSIENT OPEP.
O-OLD INJECT.
N-NEW INJECT.
CONFIG.
CONFIG.
Table
VI-12

-------
EXPERIMENTAL DATA FROM TEE
RUN
TEST
TA
TF
WA
WF
CO 2 V
C02C
02 V
02B
NOT
NO.
TYPE
o F
» F
LB/HR LB/HR
PCT
PCT
PCT
PCT
PPM
375
KSB2
400
710
91.4
2.57
3 . 9
5.9
14. 5
12.5
1.0
376
KSB2
400
710
90 . 3
2.59
4 . 5
5 .1
13.6
13.0
"l. 0
377
KSB2
405
705
48.3
1.62
6 . 8
6.4
11.1
10 .4
"1. 0
378
KSB 2
428
730
45 . 5
1 .55
6 . 8
5 . 7
10 . 8
10.5
~1. 0
379
KSB 2
375
720
47. 0
1 .50
6.4
5 .5
11. 6
10 . 5
"l. 0
380
KSB 2
412
715
45.7
1.38
5.4
5 .1
12.6
12.0
~1 . 0
381
KSB2
411
728
138.5
3.80
5.6
5 .6
12.5
11.0
"1. 0
382
KSB 2
411
721
138.7
3. 80
4.1
5 .6
14.4
12.5
"1. 0
383
KSB2
403
700
47. 1
1 .54
6 .4
5 . 7
11.3
11.5
~1. 0
384
KSB2
76
710
91.4
3.12
8.1
7 . 8
9 . 1
8.8
~1. 0
385
KSB 2
90
430
91.6
3.12
7.9
7 . 9
9.2
9 . 5
1. 0
386
KSB 2
91
470
89 . 8
2. 89
4.3
6.1
13.9
13.0
"1.0
387
KSB 2
80
740
47. 8
1.53
6.7
6.2
10 .9
9.5
"l. 0
388
KSB 2
81
740
45.5
1.48
4.3
4 . 7
13.8
11. 5
1.0
389
KSB 2
85
390
48.2
3.62
10.4
6.0
0.2
1.6
"l . 0
390
KSB 2
90
515
80.9
2.58
7.6
7 . 1
9 . 7
9.6
~1. 0
391
KSB2
93
550
80.5
2.40
2.2
5 .0
17.3
13.5
1.0
TEST_ZIEK
P-PROPANE
K-KEROSENE
E-EMISSIONS
S-STABILITY
B-BURNER
L-LOOP
1-STAND	1
2-STAND	2
RM-QQUUMZS.
NL-NORMAL LEAN
NR-NORMAL RICH
LB-LEAN BURNING
LL-LEAN LIMIT
LO- LEAN GOES OUT
RB-RICH BURNING
RL-RICH LIMIT
RO-RICH GOES OUT
G.Q.Q.&S
ERQQKMRAL_CQ_tltdElL£S
PS-STABILITY PROCED. CK.
PN-NOX EMISS. PROCED. CK.
FH-HOT SAMPLE LINE CK.
FV-VAPOR GENER. OPER. CK.
PC-FACILITY CK.
LK-LIQUID KEROSENE RUN
BURNER
PAGE 13
NOB	CO	HC
PPM	PPM	PPM
0.1	825.0	1155.
0.1	2850.0	1956.
2.7	~1.0	8.
1.4	120.0	56.
1.3	280.0	36.
0.1	860.0	760.
0.1	1025.0	690.
0.1	1570.0	1618.
1.4	695.0	625.
2.0	18.3	0.
3.4	1.0	0.
0.6	1210.0	105.
2.2	150.0	12.
0.1	2035.0	350.
27.5	102400.0	503.
5 . 8	5.0	2 .
0.2	2700.0	215.
D&.TA_Q.QtitlKtiZS
B-BAILEY N.G.
H-HC N.G.
N-NOX N.G.
V-VOLUMETRIC N.G.
F-FLOW N.G.
T-TRANSIENT OPER.
FAN	CQMKNTS
RUN PRO DAT BUR
0.0308
LL
N
0.0314
LL
N
0.0322
LB
N
0.0325
LL
N
0.0310
LL
N
0.0293
LL
N
0.0300
LL
N
0.0300
LL
N
0.0315
LB
N
0.0385
LL
N
0.0375
LB
N
0.0352
LL
N
0.0350
LL
N
0.0356
LL
N
0.0822
NR
N
0.0360
LL
N
0.0327
NL
N
MRMR-CQmENZS
OLD INJECT. CONFIG.
NEU INJECT. CONFIG.
Table
VI-13
3
5
1
7
9
0
5
9
4
7
5
1
2
1
4
6
2
0-
N-

-------
NOMENCLATURE FOR EXPERIMENTAL DATA TABLKS
Item Symbol
1,	Run No.
2.	Test type
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
TA
TF
WA
WF
C02V
C02C
02V
02B
NOT
NOB
CO
HC
FAN
Comments
Description
Chronological run sequence
Fuel type, test objectives, system
configuration, test stand
Inlet air temperature to the burner
Inlet fuel temperature to the burner
Air flow rate
Fuel flow rate
Carbon dioxide concentration as measured
by the Orsat apparatus
Carbon dioxide concentration as measured
by the Chromatograph
Oxygen concentration as measured by the
Orsat apparatus
Oxygen concentration as measured by th^
Bailey intrument
Nitrogen dioxide as measured from the
top of the generator loop
Nitrogen dioxide as measured from the
burner
Carbon monoxide as measured by the
chromatograph
Hydrocarbons as measured by the chroma-
tograph
Nominal fuel air ratio
Run comments as explained by the footing
of the data tables
Units
OK
°V
lb/hr
lb/hr
% by vo]ume
% by volume
% by volume
% bv voluni.-
PPM
PPM
PPM
PPM
Table
VI-11*

-------
SIGNIFICANT TEST PROGRAM MILESTONES
Run No.
Event
1-7
Stability test Drocedure investigation.
95
NOx line loss checkout on stand 1,
96-102
NOx line loss checkout on stand 2.
103
Started quartz tube sanrole technicue for MOx,
108
ProDane accumulator installed stand 2,
123
Air flow straightener installed stand 2.
128
Hot sample line installed stand 2,
138
Heated hydrocarbon Dump installed. Start of
good data from burner.
173
Tried to stop LB0 by raising TA. Burner out.
No data.
187
No data.
203
Liquid kerosene atomized by N2 carrier flow.
204
No data.
205-207
Vapor generator loop checkout.
208-216
Not part of this Drorram,
217
Vapor generator stack clean out.
218
Start of finalized good NOx measurement procedure.
218-220
NOx saturation evaluation.
220
NOx top of stack and burner comparison.
264-25t
New employee ran tests. Data invalid.
255
Vapor generator check out.
271
New fuel injector stand 1; start of good HC
data stand 1.
276-278
Not part of this program.
282
New fuel injector stand 2; start of good
HC data stand 2.
Table
VI-15

-------
COMPARISON OF FUEL AIR RATIO VALUES
PAGE 1
RUN NO.
FA
FA COR
1
0.0242
0.0299
2
0.0275
0.0340
3
0.0250
0.0309
4
0.0234
0.0290
5
0.0241
0 . 0298
6
0.0233
0.0288
7
0.0256
0.0317
8
0.0327
0.0405
9
0.0263
0.0325
10
0.0243
0 . 0301
11
0.0271
0.0335
12
0.0224
0.0277
13
0.0279
0.0345
14
0.0263
0.0325
15
0.0271
0.0335
16
0.0294
0.0364
17
0.0280
0.0347
18
0.0287
0.0355
19
0.0316
0.0391
20
0.0278
0.0344
21
0.0239
0.0295
22
0.0257
0.0317
23
0.0270
0.0333
24
0.0279
0.0344
25
0.0288
0.0356
26
0.0359
0.0444
27
0.0300
0.0371
28
0.0271
0.0334
29
0.0265
0.0328
30
0.0271
0.0335
31
0 .0307
0.0380
32
0.0283
0.0350
33
0.0267
0.0330
34
0.0264
0.0326
35
0.0275
0.0341
36
0.0304
0.0376
37
0.0291
0.0359
38
0.0282
0.0349
39
0 .0274
0.0339
40
0.0266
0.0329
41
0.0283
0.0350
42
0.0359
0.0444
43
0.0270
0.0334
44
0.0390
0.0482
45
0.0358
0.0443
46
0.0317
0.0392
47
0.0276
0.0341
48
0.0254
0.0314
49
0.0233
0.0288
50
0.0245
0.0303
FA -FUEL/AIR FROM FLOW METERS
FAC-FUEL/AIR FROM VOLUMETRIC CO2
FAB-FUEL/AIR FROM BAILEY DATA
FAO
FAB
FA!!
1. 0000
~1 . 0000
0 . 0299
1.0000
0.0320
0.0 340
1. 0000
0.0360
0.0309
1.0000
0.0345
0.0290
1.0000
~1.0000
0.0298
1.0000
0.0330
0.0280
1 . 0000
0.0390
0.0317
1.0000
1.0000
0. 0405
1.0000
"1.0000
0.0325
1.0000
"1 . 0000
0.0301
1.0000
"l. 0000
0.0335
1.0000
"1 . 0000
0.0277
1.0000
"l.0000
0 . 0345
1.0000
"l.0000
0.0325
0.0 331
0.0410
0.0326
1.0000
"l.0000
0.0364
1.0000
"l. 0000
0.0347
1.0000
"l.0000
0.0355
1.0000
"1.0000
0.0391
1.0000
"1.0000
0.0344
1.0000
~1.0000
0.0295
1. 00 00
"l. 0000
0.0317
1.0000
1.0000
0.0333
1.0000
"l. 0000
0.0344
0.0361
0 .0380
0.0 360
1.0000
"l.0000
0.0444
1.0 00 0
"1.0000
0.0371
1.0000
"1.0000
0.0334
1.0000
" 1 . 0000
0.0328
1. 00 00
"l. 0000
0.0335
1.0000
"1.0000
0 . 0380
1.0000
"l. 0000
0.0350
1.0000
~1 .0000
0.0330
1.0000
"1.0000
0.0326
0.0334
0 .0360
0.0332
1.cooo
"1.0000
0.0376
1.0000
"1.0000
0.0359
1.0000
"l.0000
0.0349
1.0000
1.0000
0 . 0 3 3 (J
1.0000
1.0000
0.0329
0.0331
0 .0360
0 .0350
0.0478
0.0500
0.0484
0.0328
0.0350
0.0334
1.0000
1.0 00 0
0.0482
1.0000
1.0000
0.0443
1.0000
1.0000
0.0392
1.0000
"l. 0000
0.0341
1.0000
1.0000
0.0314
1.0000
~1.0000
0.0288
0.0320
0.0330
0.0320
FACOR-CORRECTED VALUES OF FA
FAO -FUEL/AIR FROM VOLUMETRIC 02
FAN -NOMINAL FUEL/AIR FOR ANALYSES
FAC
1.0000
1.0000
1.0000
1.0000
1.0000
1.0000
1.0000
1.0000
1.0000
1.0000
1.0000
1.0000
1. 00 00
1.0000
0.0320
1.0000
1.0000
1.0000
1 . 0000
1.0000
1.0000
1.0000
1.0000
1 . 000 0
0.0360
1.0000
1.0000
1.0000
1.0000
1.0000
1.0000
1.0000
1.0000
1.0000
0.0330
1.0000
1.0000
1.0000
1.0000
1.0000
0.0320
0.0490
0.0340
1.0000
1.0000
1.0000
1.0000
1.0000
1.0000
0.0320
Table
VI-16

-------
COMPARISON OF FUEL AIR PATIO VALUES
PAGE 2
RUN NO.
FA
FA COR
51
0.0316
0.0390
52
0.0280
0.0346
53
0.0240
0.0297
54
0.0260
0.0322
55
0.1643
0.2032
56
0.1367
0.1690
57
0.1499
0.1854
58
0.1372
0.1697
59
0.1646
0.2036
60
0.1007
0.1245
61
0.1007
0.1245
62
0.1372
0.1697
63
0.1649
0.2040
64
0.1372
0.1697
65
0.1433
0.1772
66
0.1479
0.1829
67
0.1429
0.1767
68
0.1307
0.1617
69
0.1241
0.1535
70
0.1510
0.1867
71
0.1586
0. 1961
72
0.1235
0.1527
73
0.1268
0.1568
74
0.0261
0.0323
75
0.0375
0.0464
76
0.0395
0.0489
77
0.0522
0.0645
78
0.0289
0.0357
79
0.0380
0.0470
80
0.0504
0.0623
81
0.0747
0.0924
82
0.0272
0.0337
83
0.0381
0.0471
84
0.0386
0.0477
85
0.0452
0.0559
86
0. 0663
0.0820
87
0.0288
0.0356
88
0.0384
0.0475
89
0.0573
0.0709
90
0.0287
0.0355
91
0.0384
0.0475
92
0.0573
0.0709
93
0.0354
0.0388
94
0.0438
0.0479
95
0.0483
0.0528
96
0.0601
0.0658
97
0.0297
0.0325
98
0.0376
0.0412
99
0.0492
0.0539
100
0.0608
0.06C6
FA -FUEL/AIR FROM FLOW METERS
FAC-FUEL/AIR FROM VOLUMETRIC CO2
FAB-FUEL/AIR FROM BAILEY DATA
FAO
FAB
FAN
1.0000
"l.0000
0.0390
1.0000
"l.0000
0.0346
1. 0000
~1.0000
0.0297
0.0287
0.0320
0.0289
1.0000
~1. 0000
0.2032
1.0000
"l. 0000
0.1690
1.0000
"l.OOOO
0.1854
1.0000
~1.0000
0.1697
1.0000
1. 0000
0.2036
1. 0000
*1.0000
0.1245
1.0000
"l. 0000
0.1245
1.0000
"l. 0000
0. 1697
1.0000
"1.0000
0.2040
1.0000
"l.0000
0.1697
1.0000
"1.0000
0.1772
1.0000
"1.0000
0.1829
1.0000
"1.0000
0.1767
1.0000
"l.0000
0.1617
1.0000
1.0000
0.1535
1.0000
"l.0000
0.1867
1.0000
~1 . 0000
0 . 1961
1.0000
"l.0000
0 .1527
1.0000
"l.0000
0.1568
1.0000
"l.0000
0.0323
0.0396
0 . 0430
0.0 398
0.0525
0.0 510
0.0534
0.0635
0.0610
0.0637
0.0273
0.0310
0 .035?
0.0431
0.0480
0 .0440
0.0631
0.0626
0.0635
1.0000
~1.0000
0 . 0924
0.0325
0.0350
0.0323
0.0484
0.0510
0.0491
0.0475
0.0540
0.0478
0.0585
0.0590
0.0587
1.0000
"l.0000
0 . 0820
0.0343
0 . 0380
0.0345
0.0434
0.0510
0 . 0440
0 .0660
"1.0000
0 .0709
0.0367
0.0410
0.0369
0.0519
0.0550
0.0526
1.0000
"1.0000
0 .0709
0.0328
0.0380
0.0324
0.0424
0.0485
0.0432
0.0538
0.0560
0.0553
0 . 0700
"1.0000
0.0658
1.0000
~1.0000
0.0325
1.0000
"l.0000
0.0412
1.0000
~1.0000
0.0539
1.0000
"1.0000
0.0666
FACOR-CORRECTED VALUES OF FA
FAO -FUEL/AIR FROM VOLUMETRIC 01
FAN -NOMINAL FUEL/AIR FOR ANALYSES
FAC
1.0000
'l. 0000
1.0000
0.0290
1.0000
1.0000
1.0000
1.0000
1.0000
1.0000
0.0990
0.1180
1 . 0000
0.1010
0.0960
1.0000
0.0976
1.0000
0.1250
0.1075
1.0000
0.1055
1.0000
1.0000
0 .0400
0.0542
0.0640
0.0263
0.0448
0.0640
0.0840
0.0321
0.0497
0.0480
0.0590
0.0770
0.0347
0.0445
0.07 39
0.0372
0.0533
0.0839
0.0320
0.0440
0.0569
0.0663
1.0000
1.0000
1.0000
1.0000
Table
VI-17

-------
COMPARISON OF FUEL AIR FATIO VALUES
PACE 3
RUN NO.
FA
FA COR
101
0.0726
0.0795
102
0.0407
0 . 0481
103
0.0400
0. 0473
104
0.0260
0.0307
105
0.0347
0.0410
106
0.0264
0.0313
107
0.0212
0.0251
108
0.0241
0.0285
109
0.0265
0.0313
110
0.0212
0.0251
111
0.0212
0.0251
112
0.0226
0.0267
113
0.0199
0.0235
114
0.0199
0.0235
115
0.0178
0.0211
116
0.0275
0.0325
117
0.0226
0.0267
118
0.0209
0.0247
119
0.0232
0.0274
120
0.0232
0.0274
121
0.0226
0.0268
122
0.0217
0.0257
123
0.0436
0.0515
124
0.02S7
0.0304
125
0.0267
0.0316
126
0.0289
0.0342
127
0.0308
0.0364
128
0.029 3
0.0346
129
0.0274
0.0324
130
0.0315
0.0373
131
0.0372
0.0440
132
0.0571
0.0675
133
0.0305
0.0360
134
0.0381
0,0450
135
0.0235
0,0278
136
0.0288
0.0340
137
0.0381
0 . 04 50
138
0.0305
0 .0361
139
0.0393
0 .0465
140
0.0555
0,0657
141
0.0655
0 . 0774
142
0.0292
0 . 0345
143
0.0650
0.0769
144
0.0555
0.0657
145
0.0395
0.0467
146
0.0294
0.0348
147
0.0254
0.0301
148
0.0412
0.0487
149
0.0696
0.0824
150
0.0237
0.0281
PA -FUEL/AIR FROM FLOW METERS
PAC-FUEL/AIR FROM VOLUMETRIC C02
FAB-FUEL/AIR FROM BAILEY DATA
FAO
FAB
FAN
1.0000
1.0000
0.0795
1. 0000
1.0000
0.0481
1.0000
0.0550
0.0535
1. 0000
0.0350
0.0346
0 . 0589
0.0560
0.0592
0.0240
0 .0330
0.0313
1.0000
1.0000
0.0251
0.02 89
0.0272
0.0285
0.0352
0.0360
0.0356
0.020 2
0.0 2 81
0.0258
0.0228
0.0258
0.0251
0.0305
0.0311
0. 0300
0.0 243
0.0280
0. 024n
0.0252
0 . 0269
0.0243
0.0252
0.025 8
0.0248
0.0338
0.0350
0,0344
0.0275
0.0288
0 . 027G
0.0242
0.0255
0.0247
0.0298
0.0302
0.0299
0.02G9
0.0280
0.0274
0.0292
0.0298
0.0286
0.0235
~1.0000
0.0257
0.0 508
0.0519
0.0509
1.0000
0 .0405
0.0304
0.0258
"1.0000
0.0316
0.0334
~1 . 0000
0.0323
1.0000
"l.0000
0.0364
1.0000
0 .0592
0.0346
1.0000
0.0420
0.0324
1.0000
0.0375
0.0373
1.0000
"l . 0000
0. 0440
1.0000
0 . 0615
0 .0675
1.0000
0.0440
0.0360
1.0000
0.0568
0.0450
1.0000
0.0362
0.0278
1.0000
0.0420
0.0340
1.0000
0.0565
0.0450
0.0408
0.0420
0.0410
1.0000
0.0528
0.0465
1.0000
~1.0000
0.0657
1.0000
"1.0000
0.0774
1.0000
~1.0000
0.0345
1.0000
~1.0000
0.0769
1. 0000
~1.0000
0.0657
1.0000
~1. 0000
0.0467
1.0000
0.0362
0.0348
1. 00 00
0.0320
0.0301
1. 00 00
0.0492
0.0487
1.0000
0.0865
0.0824
1.0000
0.0320
0.02B1
FACOR-CORRECTED VALUES OF FA
FAO -FUEL/AIR FROM VOLUMETRIC 02
FAN -NOMINAL FUEL/AIR FOR ANALYSES
FAC
1.0000
1.0000
0.0535
0.0 346
0.0595
0.0240
1.0000
0.0277
0.0360
0.0255
0.0210
0.0295
0.025 5
0.0235
0.0245
0.0350
0.0277
0.0230
0.0300
0.0255
0.0280
0.0205
0.0510
1.0000
0.0228
0.0312
1.0000
1.0000
1.0000
1.0000
1.0000
1.0000
1.0000
1.0000
1.0000
1.0000
1.0000
0.0412
1.0000
1.0000
1.0000
1.0000
1.0000
1.0000
1.	0000
1.0000
1.0000
1.0000
0 . 0850
1.0000
Table
VI-18

-------
COMPARISON OF FUEL AIR RATIO VALVES
PAGE it
RUN NO.
FA
FACOR
FAC
FAO
FAB
FAN
151
0.0382
0.0452
"1.0000
1.0000
0.0470
0.0452
152
0.0404
0.0478
"l.0000
1.00 00
0.0502
0.0478
153
0.0254
0.0301
1.0000
1.0000
0 .0300
0.0301
154
0.0379
0.0449
"l.0000
"1.0000
0 .0465
0.0449
155
0.0273
0.0323
1.0000
1.0000
0.0345
0.0323
156
0.0232
0.0275
1.0000
"1.0000
0 . 0265
0.0275
157
0.0317
0.0375
"1.0000
"1.0000
0.0380
0.0375
158
0.0345
0.0408
"1 . 0000
"1.0000
0.0420
0 .0408
159
0.0411
0.0486
~1.0000
"1.0000
0.0525
0.0486
160
0.0662
0.0782
"1. 0000
"l.0000
0.0775
0 . 0782
161
0.0672
0.0795
~1.0000
"1.0000
0.0750
0.0795
162
0 . 0844
0.0998
1 . 00 00
1.0000
0.0930
0.0998
163
0.0296
0.0350
"l.00 00
"l.0000
0. 0 39 0
0.0350
16 4
0.0409
0.0484
"1.0000
"1.0000
0 .0595
0.0484
165
0.0429
0.0507
"1.00 00
"1.0000
0.0485
0.0507
166
0.0359
0.0425
"l.0000
"l.0000
0.0415
0.0425
167
0.0455
0.0538
"l.0000
"l.0000
0.0495
0.0538
168
0.0526
0.0623
"1.0000
"1.0000
0.0578
0.0623
169
0.0395
0.0467
~1.0000
"1.0000
0.0 44 5
0.0467
170
0.0312
0.0369
"l.0000
"l.0000
0.0 360
0 .0369
171
0.0279
0.0330
"1.0000
"l.0000
0.0320
0.0330
172
0.0223
0.0264
"l.0000
"1.0000
0.0258
0 . 0264
174
0.0352
0.0416
"l.0000
"1.0000
0 .0400
0 . 0416
175
0.0429
0.0507
"l.0000
"l.0000
0.0500
0.0507
176
0.0472
0.0559
"1.0000
"1.0000
0.0535
0.0559
177
0.0243
0.0288
0.0270
0.0273
0.0282
0.0271
178
0.0293
0.0346
"1.0000
"l.0000
0 . 0345
0.0346
179
0.0454
0.0537
"1.0000
"l.0000
0.0535
0.0537
180
0.0386
0.04 5 7
"l.0000
"1.00 00
0.0458
0.0457
181
0.0336
0.0397
"1.0000
"1.0000
0.0400
0 . 0397
182
0.0295
0.0349
"l. 00 00
~1.0000
0.0360
0.0349
183
0.0265
0.0313
"l.0000
"1.0000
0.0315
0.0313
184
0.0240
0.0284
"l. 00 00
~1.0000
0.0290
0.0284
185
0.0271
0.0321
"1.0000
"1.00 00
0.0325
0.0321
186
0.0304
0.0359
0.0360
0.0361
0.0385
0.0360
188
0.0424
0.0464
0.0450
0.0461
0.0510
0.0455
189
0.0351
0.0384
0.0360
0 .0378
0.0390
0.0369
190
0.0297
0.0326
'1.0000
"1.0000
0.0320
0.0326
191
0.0351
0.0384
"1.0000
"1.0000
0.0370
0.0384
192
0.0420
0.0460
"l.0000
"l.0000
0 . 0445
0.0460
193
0.0766
0.0839
1.0000
"1.0000
0.0900
0.0839
194
0.0334
0.0365
0.0360
0 . 0368
0 . 040 0
0.0364
195
0.0296
0.0324
"1.0000
"1.00 00
0.0370
0.0324
196
0.0266
0.0292
0.0328
"1.0000
0.0328
0.0328
197
0.0303
0.0331
0.0345
0.0362
0.0367
0 . 0354"
198
0.0401
0.0439
"l.0000
"1.0000
0.0405
0.0439
199
0.0598
0.0655
~1.0000
"1.0000
0.0630
0.0655
200
0.0403
0.0441
0.0305
0 .0384
0.0415
0.0441
FA -FUEL/AIR FROM FLOW METERS
FAC-FUEL/AIR FROM VOLUMETRIC C01
FAB-FUEL/AIR FROM BAILEY DATA
FACOR-CORRECTED VALUES OF FA
FAO -FUEL/AIR FROM VOLUMETRIC 02
FAN -NOMINAL FUEL/AIR FOR ANALYSES
Table
VI-19

-------
COMPARISON OF FUEL AIR PATIO VALUES
PAGE 5
RUN no.
201
202
203
FA
0. 0445
0.0389
0.0378
FA COR
0.0526
0.0460
0.0414
FAC
1. 0 00 0
1.0000
1.0000
FAO
1.0000
'l. 0000
1.0000
F>\b
1.0000
1 . 0000
1.0000
FAI'
0 . 0 5 ? 3
0 .0460
0.0414
205
0.0378
0.0447
~1.0000
1.0000
0.0464
0.0447
206
0.0352
0.0416
"l.0000
1.0000
0.0435
0.0416
207
0.0303
0.0358
1.oooo
1.00 00
0.0347
0.0358
217
0.0327
0.0387
"1.0000
"1.0000
0.0317
0.0387
218
0.0251
0.0249
0.0440
0.0352
"l.0000
0.0396
219
0.0372
0.0368
0.0520
0 .0484
~1.0000
0. 0502
220
0.0317
0 .0375
"l.0000
"1.0000
0 . 0406
0.0375
221
0.0312
0.0369
"l.0000
"l.0000
0 . 0378
0.0369
222
0.0276
0.0327
"l.OOOO
"1.0000
0.0346
0.0327
223
0.0336
0.0397
"l.0000
"1.0000
0.0400
0.0397
224
0.0253
0.0300
"1.0000
"l.0000
0 .0378
0 .0356
225
0.0281
0 . 0333
"l.0000
~1 . 0000
0.0400
0 .0395
226
0.0240
0.0284
1.0000
"1.0000
0.0346
0.0337
227
0.0314
0.0371
"1.0000
"l.0000
0.0420
0.0371
228
0.0312
0.0369
"1.0000
"1.0000
0.0315
0.0369
229
0.0312
0.0369
"l.0000
"1.0000
0.0378
0.0369
230
0.0313
0.0370
"l.0000
"1.0000
0 .0 360
0.0370
231
0.0296
0.0350
~1. 0000
"l. 00 00
0.0310
0.0350
232
0.0187
0.0221
"1.0000
"1.0000
0.0333
0.0333
233
0.0139
0.0164
"1.0000
~1 . 00 00
0.0395
0 .0305
234
0.0364
0.0430
~1.0000
"1. 0000
0.0522
0.0430
235
0.0321
0.0380
~1.0000
"1.0000
0.0460
0.0380
236
0.0264
0.0312
0. 0440
0.0328
0.0380
0.0384
237
0.0344
0.0407
"1.0000
"1.0000
0.0522
0.0407
238
0.0316
0.0312
0.0270
0.0275
0.0350
0.0312
239
0.0388
0.0384
0.0350
0.0340
0.0410
0 .0345
240
0.0417
0.0413
0.0420
0.0420
0.0465
0.0420
241
0.0512
0.0507
0.0465
0.0465
0.0520
0.0465
242
0.0498
0.0492
0.0475
0.0500
0.0520
0.0488
243
0.0537
0.0531
0.0575
0.0570
0.0580
0.0573
244
0. 0342
0.0339
0.0320
0.0325
0.0350
0.0323
245
0.0398
0.0393
0.0380
0.0387
0.0410
0.0384
246
0.0414
0.0409
J.0000
"l.0000
0.0289
0. 0409
247
"l.0000
"l. 000 0
~1.0000
" 1. 0000
0.0295
0.0295
248
"l.0000
"1 . 0000
"1.0000
1.0000
0.0360
0.0360
249
0.0548
0.0542
"1.0000
~1.0000
0.0360
0.0542
250
0.0553
0.0548
1.0000
1.0000
0.0490
0 .0548
FA -FUEL/AIR FROM FLOW METERS
FAC-FUEL/AIR FROM VOLUMETRIC C02
FAB-FUEL/AIR FROM BAILEY DATA
FACOR-CORPECTED VALUES OF FA
FAO -FUEL/AIR FROM VOLUMETRIC 02
FAN -NOMINAL FUEL/AIR FOR ANALYSES
Table
VI-20

-------
COMPARISON OF FUEL AIR RATIO VALUES
PAGE 6
RUN NO.
FA
FACOR
FAC
FAO
FAB
FAN
251
0.0322
0.0319
1.0000
"l.0000
0.0348
0.0319
252
0.0324
0.0320
"1.0000
"l.0000
0 . 0360
0.0320
253
0.0373
0.0369
"1.0000
"l.0000
0.0410
0.0369
254
"l.0000
"l.0000
0.0410
"1.0000
0 .0470
0.0410
255
0.0242
"1.0000
0.0388
"1.0000
"1.0000
0.0388
256
"l.OOOO
"1.0000
0.0460
0.0460
0.0460
0 . 0460
257
"1.0000
"l.OOOO
0.0505
0.0500
0.0500
0 .0503
258
0.0539
0.0533
0.0589
0 .0564
0.0565
0.0576
259
"1.0000
"1.0000
0.0640
"1.0000
1.0000
0 .0040
260
0.0627
0.0620
0.0580
0.0650
0.0625
0 .0615
261
"1.0000
"l.OOOO
0.0760
1.0000
"l.0000
0.0760
262
0.0335
0.0331
0.0330
0.0331
0.0348
0.0331
263
0.0401
0.0397
0.0390
0.0372
0.0410
0 .0381
264
0.0344
0.0341
0.0320
0 .0314
0.0310
0 .0317
265
"1.0000
"l.OOOO
0.0375
0.0370
0 .0375
0.0372
266
0.0337
0.0334
0.0360
0 .0367
0 .0383
0.0363
267
0.0327
0.0324
0.0335
0.0311
0.0323
0.0323
26 8
0.0365
0.0361
0.0375
0.0370
0.0380
0.0372
269
0.0427
0.0422
0.0420
0.0422
0.0438
0.0421
270
0.0443
0.0438
0.0435
0.0443
0.0467
0.0439
271
0.0354
0.0350
0.0350
0 .0364
0.0365
0.0357
272
0.0420
0.0415
0.0410
0.0437
0.0455
0.0424
273
0.0465
0.0460
0.0435
0 .0428
0.0480
0.0432
274
0.0460
0.0455
0.0490
0.0478
0.0495
0.0484
275
0.0508
0.0503
0.0537
0.0519
0 .0540
0.0528
279
0.0310
0.0366
0.0360
0.0381
0.0400
0.0371
280
0.0368
0.0436
"1.0000
"1.0000
"1.0000
0.0436
281
0.0311
0.0368
0.0375
0 .0384
0 .0395
0.0380
282
0.0291
0.0344
0.0390
0.0387
0.0410
0.0389
283
0.0264
0.0313
0.0425
0.0346
0.0360
0.0386
284
0.0232
0.0275
0.0270
0.0273
0.0310
0.0271
285
0.0242
0.0286
0.0310
0.0314
0 . 0315
0.0312
286
0.0332
0.0328
0.0310
0.0331
0.0360
0.0321
287
0.0340
0.0337
0.0280
0.0299
0.0335
0 .0337
288
0.0351
0.0348
0.0360
0.0364
0.0370
0.0362
289
0.0382
0.0378
0.0390
0.0380
0.0385
0.0385
290
0.0441
0.0436
0.0460
0.0461
0.0478
0.0460
291
0.0780
0.0772
"1.0000
"l.0000
0.0895
0.0895
292
0.0815
0.0006
"l.0000
"1.0000
0.1000
0 .1000
293
0.0818
0.0809
0.0B70
1.0000
0.0945
0 .0870
294
0.0675
0.0668
0.0805
1.0000
0.0770
0.0805
295
0.0782
0.0774
0.0845
1.0000
0.0875
0 .0845
2 96
0.0751
0.0743
0.0860
1.0000
0.0820
0.0860
297
0.0890
0.0880
0.0905
1.0000
0.0960
0.0905
298
0.0774
0.0766
0.0820
1.0000
0.0790
0.0820
299
0.0918
0.0908
1.0000
1.0000
0.0880
0.0880
300
0.0874
0.0865
1.0000
"1.0000
0.0750
0.0750
FA -FUEL/AIR FROM FLOW METERS	FACOR-CORRECTED VALUES OF PA
FAC-FUEL/AIR FROM VOLUMETRIC CO2 FAO -FUEL/AIR FROM VOLUMETRIC 02
FAB-FUEL!AIR FROM BAILEY DATA	PAN -NOMINAL FUEL/AIR FOR ANALYSES
Table
VI-21

-------
COMPARISON OF FUEL AIR RATIO VALUES
PAGE 7
RUN NO.
FA
FACOR
301
0.0701
0.0694
302
0 . 09 03
0.0894
303
0.0741
0.0733
3 04
0.0772
0.0764
305
0.0938
0.0928
306
0.0734
0 .0726
307
0.0679
0.0672
308
0.0343
0.0406
309
0.0316
0.0373
310
0.0186
0.0220
311
0.0354
0.0419
312
0.0269
0.0318
313
0.0212
0.0251
314
0.0315
0.0373
315
0.0258
0.0305
316
0.0239
0.0282
317
0.0217
0.0256
318
0.0274
0.0324
319
0.0225
0.0266
320
0.0332
.0. 039 3
321
0.0329
0.0389
322
0.0260
0.0308
323
0.0242
0.0287
3 24
0.0328
0.0388
325
0.0280
0.0332
326
0.0260
0 . 0307
327
0.0311
0.0368
328
0.0243
0.0288
329
0.0185
0.0219
330
0.0324
0.0384
331
0.0248
0.0293
332
0.0226
0.0267
333
0.0238
0.0282
334
0.0206
0.0244
335
0.0301
0.0330
336
0.0260
0.0285
337
0.0382
0.0418
338
0.0359
0.0393
339
0.0326
0.0357
340
0.0274
0.0324
341
0.0276
0.0326
342
0.0309
0.0338
343
0.0278
0.0329
344
0.0245
0.0289
345
0.0294
0.0347
346
0.0264
0.0312
347
0.0357
0 .0422
348
0.0319
0.0350
349
0.0283
0.0310
350
0.0293
0.0321
FA -FUEL/AIR FROM FLOW METERS
FAC-FUEL/AIR FROM VOLUMETRIC COl
FAB-FUEL/AIR FROM BAILEY DATA
FAO
FAP
fa>;
1 . 00 00
0.0730
0 .0780
1.0000
0.0930
0 .0910
1.00 00
0.0805
0.0820
1 . 0000
0.0845
0.0860
1.0000
0.0895
0.0880
1 . 0000
0.0755
0.0750
1.0000
0.0745
0.0765
0.0360
0 . 0390
0.0350
0.0317
0.0338
0.0323
0.0193
0 . 0275
0.0220
0.0378
0.0390
0.0377
0.0323
0 .0330
0.0326
0.0255
0.0272
0 .0252
0.0340
0 . 0390
0.0345
0 . 030 5
0.0335
0.030 2
0.0325
0 . 0339
0.0323
0.0270
0 .0279
0 .0260
0.0317
0.0338
0.0316
0.0264
0.0279
0.0266
0.0446
0 .0390
0 .0415
0.039 0
0.0395
0.0383
0.0317
0.0338
0.0311
0. 02 81
0 . 02 90
0.0287
0.0378
0.0390
0.0374
0.0314
0. 0340
0.0332
0.0287
0 . 0317
0 .0307
0 . 0390
0.0395
0.0385
0 .0299
0 .0320
0. 029''
0 . 0249
0 . 0270
0.0245
0.0387
0 .0396
0.0331
0.0314
0 .0330
0 .0309
0.0270
0.0320
0.0260
0.0370
0 . 0330
0.0282
0.0252
0 .0280
0.0246
0.0361
0 . 0390
0.0370
0.0322
0.0335
0.0311
0.0375
0.0400
0.0362
0.0362
0.0380
0.0393
0.0393
0.0390
0.0379
0 .0320
0 . 0388
0.0320
0.0331
0.0388
0.0336
0.0375
0 . 0390
0.0365
0.0345
0.0335
0.0337
0.0 30 5
0.0305
0.0300
0.0352
0.0352
0.0348
0.0317
0.0310
0.0313
0.0420
0.0390
0.0417
0.0393
0 . 0360
0.0401
0.0350
0.0317
0.0342
0.0375
0.0335
0.0380
FACOR-CORRECTED VALUES OF FA
FAO -FUEL/AIR FROM VOLUMETRIC 02
FAN -NOMINAL FUEL/AIP FOR ANALYSES
FAC
0.0780
0.0910
0.0820
0.0860
0.OOOO
0.0750
0.0765
0.0340
0.0330
0.0205
0.0375
0.0330
0.0250
0.0350
0.0300
0.0320
0.0250
0.0315
0 . 0240
0.0384
0.0375
0.0305
0.0260
0.0370
0.0280
0.0 2 80
0.0380
0.0295
0.0 240
0. 0375
0.0305
0.0250
0. 02 85
0.0240
0.0380
0.0300
0.0350
1.0000
0.0365
0.0320
0.0340
0.0355
0.0330
0.0295
0.0345
0.0310
0.0415
0.0410
0.0335
0.0385
Table
vi-22

-------
COMPARISON OF FUEL AIR RATIO VALUES
PAGE 8
RUN NO.
FA
FACOR

351
0.0325
0.0355
0.
352
0.0285
0.0312
0 .
353
0.0323
0.0354
0.
354
0.0322
0.0352
0.
355
0.0262
0.0287
0.
356
0.0388
0.0425
0.
357
0.0275
0.0302
0 .
3 58
0.0350
0.0383
0.
359
0.0302
0.0331
0.
360
0.0324
0.0355
0 .
361
0.0292
0.0319
0.
362
0 .0367
0.0402
0.
363
0.0336
0.0368
0.
3 64
0.0390
0.0427
0 .
365
0.0438
0.0479
0.
366
0.0324
0.0355
0.
367
0.0292
0.0320
0.
368
0.0254
0.0278
0.
369
0.0320
0.0350
0 .
370
0.0328
0.0359
0.
371
0.0318
0.0348
0.
372
0.0308
0.0337
0.
373
0.0299
0.0328
0.
374
0.0290
0.0318
0.
375
0.0281
0.0308
0.
376
0.0287
0.0314
0.
377
0.0335
0.0367
0.
378
0.0341
0.0373
0.
379
0.0319
0.0350
0.
380
0.0302
0.0330
0.
381
0.0274
0.0300
0.
382
0.0274
0.0300
0.
383
0.0327
0.0358
0.
384
0.0341
0.0374
0.
385
0.0340
0.0373
0.
386
0.0322
0.0352
0.
387
0.0320
0.0350
0.
388
0.0325
0.0356
0.
389
0.0751
0.0822
0.
390
0.0319
0.0349
0.
391
0.0298
0.0327
0.
FA -FUEL/AIR FROM FLOW METERS
FAC-FUEL/AIR FROM VOLUMETRIC CO2
FAB-FUEL/AIR FROM BAILEY DATA
FAO
FAB
FAN
0 . 0411
0. 0415
0.0421
0.0331
0.0370
0.0336
0.0411
0.0415
0.0421
0.0362
0.0380
0.0359
0.0295
0.0320
0.0287
0.0378
0.0410
0.0371
0.0344
0.0370
0.0337
0.0313
0.0410
0.0312
0.0335
0.0415
0.0335
1.0000
0.0400
0.0355
0.0313
0.0350
0.0312
0.0421
0.0430
0.0420
0.0335
0.0330
0.0332
0.0384
0.0330
0.0379
0.0410
0.0370
0.0416
0.0295
0.0250
0.0287
0.0350
0.0280
0.0340
0.0295
0.0250
0.0287
0.0 384
0.0410
0.0379
0.0365
0.0370
0.0365
0.0385
0.0370
0.0383
0.0375
0.0360
0.0362
0.0360
0.0370
0.0355
0.0335
0.0330
0 .0327
0.0230
0.0260
0.0308
0.0265
0.0260
0.0314
0.0325
0.0340
0.0322
0.0330
0.0340
0.0325
0.0310
0.0330
0.0310
0.0280
0.0290
0.0330
0.0280
0.0315
0.0300
0.0230
0.0280
0.0300
0.0320
0.0300
0.0315
0.0390
0.0390
0.0385
0.0390
0.0375
0.0375
0.0 240
0.0260
0 .0352
0.0330
0.0370
0.0350
0.0250
0.0300
0.0356
1.0000
0.0770
0.0822
0.0370
0.0370
0.0360
0.0140
0.0260
0.0327
-CORRECTED VALUES OF FA
-FUEL/AIR FROM VOLUMETRIC 02
-NOMINAL FUEL/AIR FOR ANALYSES
FAC
0 43 0
0340
0430
0355
0 2 00
0365
0330
0310
0335
0350
0310
0420
0330
0375
0415
0280
0330
0280
0375
0355
0380
0350
0 3 50
0320
0190
0220
0320
0320
0310
0260
02 GO
0200
0310
0380
0360
0210
0320
0210
0780
0350
0115
FACOR
FAO
FAN
Table
VI-23

-------
Kind, AIH HATIO CORRECTION FACTORS
FOK FLOWMSTKH DATA
RUN CHODP
(i^top wbich
factor obtained)
KUKL
TKST
STAND
KACTOK
COhKiNT
t-larly runs
(Approx. runs
IS to 01)
Propane
1
0.80ft'7 ± .0^7
Air leakc
later runs
(Approx. runs
?UU to POOl
Propane
1
1.01? + .056«
mr leaks
ellmina ted
Addtoy. runs
i?3 to 3^1
ProDane
?
O.BU^ + .0560
Air leaks
Approx, runs
9^ to 369
Kerosene
2
0.9HT ± . 102fi
Mr leaks
NOTKi To correct measured -flowmeter fuel-a1r valuer, divide by factor.
iable
VI-2U

-------
THEORETICAL FLAME TEMPERATURES	PACE 1
RUN
AIR TEMP
FUEL/AIR
FLAME Tl
NO.
OF

OF
1
75
0.0299
2157
2
80
0.0340
2400
3
85
0.0309
2221
4
85
0.0290
2105
5
85
0.0298
2154
6
88
0.0288
2099
7
90
0.0317
2272
8
90
0.0405
2776
9
90
0.0325
2320
10
90
0.0301
2177
11
90
0.0335
2378
12
90
0.0277
2031
13
90
0.0345
2436
14
90
0.0325
2320
IS
90
0.0326
2325
16
90
0.0364
2545
17
90
0.0347
2447
IB
90
0.0355
2492
19
91
0.0391
2699
20
92
0.0344
24 34
21
9 2
0.0295
2144
22
9 3
0.0317
22 79
23
93
0.033 3
2372
24
93
0.0344
2436
25
93
0.0360
2526
26
90
0.0444
2982
27
93
0.0371
2585
28
96
0.0334
2381
29
97
0.0328
2342
30
97
0.0335
2 3 84
31
102
0.0380
2638
32
105
0.0350
2475
33
108
0 .0330
2364
34
95
0.0326
2333
35
102
0.0332
2372
36
112
0.0376
2627
37
117
0.0359
2536
38
120
0.0349
2480
39
120
0.0339
2425
MO
120
0.0329
2364
41
124
0.0326
2350
42
91
0.0484
3167
43
93
0.0334
2377
44
87
0.0482
3154
45
85
0.0443
2973
46
84
0.0392
2701
47
82
0.0341
2410
48
82
0.0 314
2252
49
80
0.02 88
2091
50
80
0.0320
2283
Table
VI-25

-------
THEORETICAL FLAME TEMPERATURES
PAGE 2
RUN	AIR TEMP
NO.	OF
51	83
52	82
53	82
54	83
55	83
56	81
57	82
58	85
59	85
60	90
61	90
62	85
63	87
b4	85
65	88
66	85
67	85
68	70
69	82
70	85
71	86
72	90
73	9 3
74	86
75	85
76	85
77	85
78	90
79	92
80	90
81	92
82	103
83	105
84	100
85	103
86	105
87	300
88	300
89	300
90	300
91	310
92	300
93	250
94	250
95	250
96	250
97	95
98	95
99	95
100	100
FUEL/AIR FLAME TEMP
OF
0.0390
2688
0.0346
2437
0.0297
2150
0.02 89
2098
0.2032
1492
0.1690
1708
0.1854
1639
0.1697
1794
0.2036
1490
0.1245
2447
0.1245
2447
0.1697
1794
0.2040
1488
0.1697
1794
0.1772
1717
0.1829
1661
0.1767
1720
0.1617
1875
0.1535
1978
0.1867
1629
0.1961
15 49
0.1527
1933
0.1568
1942
0.0323
2306
0.0464
3075
0.0534
3356
0.0637
3554
0.0357
2504
0.0440
2962
0.0635
3555
0.0924
3143
0.0323
2321
0.0491
3203
0 . 0478
3143
0.0587
3502
0.0820
3371
0.0345
2587
0.0440
30 80
0.0709
3658
0.0369
2717
0.0526
3446
0.0709
3658
0.0324
2362
0. 04 32
29 26
0.05 53
3431
0.0658
3634
0.0325
2251
0.0412
2724
0.0539
330 5
0.0666
3563
Table
VI-26

-------
THEORETICAL FLAME TEMPERATURES	PAGE 3
RUN
AIR TEMT
FUEL/AIR
FLAME Tl
NO.
OF

OF
101
102
0.0795
3485
102
102
0.0481
3160
103
400
0.0473
3284
104
400
0.0307
2459
105
84
0.0410
2801
106
85
0.0313
2246
107
85
0.0251
1872
108
85
0.0285
2076
109
85
0.0313
2248
110
400
0.0251
2131
111
400
0.0251
2131
112
400
0.0267
2229
113
400
0.0235
2037
114
400
0.0235
20 39
115
400
0.0211
1885
116
400
0.0325
2555
117
400
0.0267
2229
118
400
0.0247
2110
119
400
0 . 0274
2272
120
400
0.0274
2269
121
400
0.0268
2234
122
400
0.0257
2167
123
74
0.0509
3262
124
80
0.0304
2187
125
85
0.0316
2262
126
95
0.0342
2423
127
95
0.0364
2548
128
90
0.0346
2445
129
80
0.0324
2308
130
80
0.0373
2586
131
80
0.0440
2955
132
80
0.0675
3562
133
80
0.0360
2517
134
80
0.0450
3007
135
80
0.0278
2033
136
88
0.0340
2409
137
90
0.0450
3013
138
95
0.0410
2807
139
95
0.0465
30 82
140
95
0.0657
3569
141
95
0.0774
3460
142
95
0.0345
2440
143
95
0.0769
3472
144
95
0.0657
3569
145
95
0.0467
30 95
146
95
0.0348
2456
147
95
0.0301
2182
148
95
0.0487
3183
149
95
0.0824
3356
150
95
0.0281
2061
Table
VI-27

-------
THEORETICAL FLAME TEMPERATURES	PACE 4
RUN
AIR TEMP
FUEL/AIR
FLAME Ti
110.
OF

OF
151
400
0.0452
3196
152
400
0.0478
3303
153
360
0.0301
2391
154
355
0.0449
3155
155
270
0.0323
2443
156
260
0.0275
2158
157
260
0.0375
2718
158
260
0.04 0 8
2898
159
265
0.0486
3268
160
270
0.0782
3546
161
275
0.0795
3525
162
275
0.0998
3096
163
360
0.0350
2657
164
370
0.0484
3315
165
375
0.0507
3412
166
365
0.0425
3050
167
370
0.0538
3516
168
370
0.0623
3687
169
345
0.0467
3229
170
325
0.0369
2734
171
305
0.0330
2511
172
275
0.0264
2105
173
"1
"l. 000 0
~1
174
400
0.0416
3030
175
405
0.0507
3429
176
405
0.05 59
3 59 5
177
405
0.0288
2354
178
405
0.0346
2675
179
350
0.0537
3502
180
345
0.0457
3185
181
340
0.0397
2891
182
340
0.0349
2639
183
335
0.0313
2439
184
334
0.0284
2271
185
335
0.0321
2481
186
335
0.0359
2688
187
~1
"1.0000
"l
188
430
0.0455
3147
189
430
0.0369
2734
190
4 30
0.0326
2513
191
400
0.0384
2788
192
405
0.0460
3151
193
410
0.0839
3587
194
410
0.0364
2695
195
410
0.0324
2487
196
410
0.0292
2313
197
430
0.0354
2657
198
360
0.0439
30 2 7
199
340
0.0655
3677
200
410
0.0441
3070
Table
VI-28

-------
THEORETICAL FLAME TEMPERATURES	PACE 5
RUN
AIR TEMP
FUEL/AIR
FLAME T
HO.
OF

OF
201
86
0.0526
3331
202
82
0.0460
30 54
203
110
0.0414
2749
204
"1
~1.0000
"1
205
313
0.0447
3120
206
253
0.0416
2935
207
245
0.0358
2621
208
"l
~1.0000
"l
209

"1.0000
"l
210
"l
~1.0000
"1
211
"1
~1.0000
"1
212
~1
"1.0000
"l
213
~1
~1.0000
"l
214
~1
~1.0000
"1
215
~1
~1.0000
"1
216
"1
"1.0000
"l
217
80
0.0387
2666
218
100
0.0311
2245
219
110
0.0460
3072
220
100
0.0375
2611
221
100
0.0369
2578
222
100
0.0327
2339
223
100
0.0397
2738
224
100
0.0300
2179
225
100
0.0333
2374
226
100
0.0284
2083
227
76
0.0371
2574
228
82
0.0369
2566
229
85
0.0369
2572
230
433
0 .0370
2817
231
420
0.0350
2706
232
410
0.0221
1959
233
407
0.0164
1584
234
90
0.0430
2914
235
92
0.0380
2632
236
89
0.0312
2244
237
82
0.0407
2781
238
79
0.0390
2686
239
80
0.0480
3143
240
82
0.0516
3293
24 1
82
0.0634
3550
242
78
0.0615
3530
243
80
0.0664
3563
244
79
0.0323
2300
245
80
0.0384
2647
246
80
0.0511
3275
247
75
0.0295
2131
248
85
0.0360
2519
249
75
0.0678
3558
250
85
0.0684
3559
Table
VI-?q

-------
THEOPETICAL FLAME TEMPEPATURES	PAGE 6
RUN
AIR TEMP
FUEL/AIR
FLAME Tl
NO.
OF

OF
251
85
0 . 0399
2738
252
90
0 .0400
2751
253
92
0.0462
3068
254
93
0.0410
2809
255
93
0 .0388
2681
256
85
0 . 0460
3056
257
85
0.0503
3242
258
85
0.0576
3472
259
85
0.0640
3556
260
85
0.0775
34 5 3
261
89
0 . 0760
3482
262
90
0.0331
2354
263
97
0.0381
2645
261
85
0.0317
2270
265
85
0.0372
2589
266
96
0 . 0363
2545
267
96
0.0323
2313
268
96
0.0372
2595
26 9
96
0.0421
2869
270
96
0.0439
2961
271
100
0.0357
2511
272
100
0 . 0424
2884
273
100
0.0432
2926
274
100
0.0484
3172
275
95
0.0528
3343
276
"l
"1.0000
1
277
"l
~1.0000
"l
278
"1
~1.0000
"l
279
109
0.0371
2594
280
110
0.0430
2953
281
110
0.0380
2644
282
94
0.0389
2685
283
100
0.0313
2257
284
105
0.0271
2011
285
116
0.0312
2263
286
82
0.0321
2290
287
92
0.0290
2111
288
90
0.0362
2533
289
95
0.0305
2664
290
99
0.0460
3065
291
73
0.0965
3042
292
75
0.1008
2948
293
82
0.1011
2945
294
90
0.0835
3329
295
72
0.0967
3036
296
86
0.0929
3128
297
92
0.1100
2754
298
95
0.0957
3073
299
96
0.1135
2682
300
85
0.1081
2792
Table
VI-30

-------
THEORETICAL FLAME TEMPERATURES	PAGE 7
RUN
AIR TEMP
FUEL/AIR
FLAME Ti
HO.
OF

OF
301
88
0.0 867
3261
302
92
0.1117
2718
303
74
0.0917
3147
SOU
87
0.0954
3074
305
100
0.1160
2631
306
100
0 .0907
3183
307
104
0 . 0839
3329
308
90
0.0406
2782
309
98
0.0323
2317
310
94
0.0199
1554
311
385
0.0377
2816
312
388
0.0326
2553
313
395
0.0252
2138
314
340
0.0345
2616
315
364
0.0302
2402
316
400
0.0323
2 544
317
435
0.0260
2215
318
350
0 .0316
2465
319
350
0.0252
2097
320
351
0.0393
2878
321
140
0.0383
2680
322
115
0.0311
2257
323
105
0.0271
200 8
321
95
0 .0374
2605
325
105
0.0332
2371
326
110
0.0284
2090
327
412
0.0385
2878
328
395
0.0297
2397
329
390
0.0245
2086
330
407
0.0381
2854
331
400
0.0309
2470
332
435
0.0260
2215
333
392
0.0282
2307
3 34
395
0.0246
2099
335
92
0.0370
2499
336
90
0.0311
2165
337
90
0.0362
2453
338
90
0.0393
2620
339
100
0.0379
2550
340
395
0.0320
2524
341
150
0.0336
2427
342
90
0.0365
2467
343
102
0.0329
2353
344
100
0.0289
2117
345
102
0.0347
2458
346
105
0.0312
2259
347
105
0.0422
2880
348
107
0.0401
2675
349
100
0.0342
2351
350
101
0.0380
2555
Table
VI-31

-------
THEORETICAL FLAME TEMPERATURES	PAGE 8
RUN
AIR TEMP
FUEL/AIP
FLAME TE
no.
OF

OF
351
105
0.0421
2780
352
110
0.0336
2320
353
120
0.0421
2789
354
430
0.0359
2683
355
438
0.0287
2312
356
450
0.0371
2761
357
140
0.0337
2349
358
110
0.0312
2183
359
120
0.0335
2322
360
380
0.0355
2624
361
40 0
0.0312
2413
362
405
0.0420
2969
363
405
0.0332
2527
364
405
0.0379
2767
365
405
0.0416
2949
366
410
0.0287
2289
367
405
0.0340
2567
368
410
0.0287
2289
369
400
0.0379
2703
370
403
0.0359
2664
371
426
0.0348
2627
372
425
0.0337
2568
373
425
0.0328
2518
374
430
0.0318
2470
375
400
0.0308
2393
376
40 0
0.0314
2426
377
405
0.0367
2707
378
428
0.0373
2754
379
375
0.0350
2594
380
412
0.0330
2523
381
411
0.0300
2362
382
411
0.0300
2360
383
403
0.0358
2660
3 8 4
76
0.0374
2506
385
90
0.0373
2510
386
91
0.0352
2399
387
80
0.0350
2380
388
81
0.0356
2411
389
85
0.0822
3423
390
90
0.0349
2380
391
93
0.0327
2256'
Table
VI-32

-------
FUEL\_-Av\P. COMBUSTION DKTK
OCTKHE.
PKOPfc^E.
S
I
0
ti
X
1
•a
I /'
NO
TOTAL *
COttbUS>-Vtt>V-tS>
,\
K Oz

\
'CO / /
S

\
\

7 Hi
i
\ / / *	ViVUKT P%>OV%.*>
I K&JxbS'. 06}C0nftlHi KS
' TOTKL COMbOSTIft\.tt5
V
' \\
£ .S TDJCH fX>AA*tJHkHM 0.0*40
FO/r octahb aotts
' ^—
1
I	I
0.02
0.0U 0.06 0.08 0.10 0.12 0.14 0.16
FUEL AIR RATION f/a
Reference! Conbmtion of Hydrocarbons— Property Tables
Purdue University, Eng. Ext. Ser #122, Hay 66
Figure
71-1

-------
COMPARISON OF VOLUMETRIC AND BAILEY OXYGEN DATA
VOLUMETRIC OXYGEN DATA	02V
0.00	14.00	8.00 12.00 16.00 20.00

-------
0.00
THEORETICAL C02
PROPANE AIR COMBUSTION
OCTANE AIR COHBUSTION
CARBON DIOXIDE VALUES VERSUS NOMINAL FUEL/AIR RATIO
0.02
0.0U	0.06	0.08
NOMINAL FUEL/AIR RATIO
0.1°
0.12
0.1U
FAN
f
-THEORETICAL 0X1GEH FOR
PROPANE AIR COMBUSTION
-r
VOLUMETRIC 0JCT3DJ DATA VERSUS NOMINAL FUEL/AIR RATIO
0.16
0.02	0,0*
0.06
0.08
0.10	0.12	0.1*	0.16
NOMINAL FUEL/AIR RATIO

-------
FLOW METER FUEL/AIP PATIO	FA

-------
CARBON DI OX I OF DATA FROM THE GAS CHROMATOGRAPH
0.00
0.02
0.0U	0.06-	O.OB	0.10	0.12
NOMINAL FUEL/AIR RATIO	FAN
0.1U
0.16
C02C
0.00
2.00	It.00	6.00	B.00 10.00 12.00
C02 DATA FROM THE VOLUMETRIC ANALYZER - C02V %
XI
«	COMPARISON OF CARBON DIOXIDE CHROMATOGRAPH AND VOIUMETPIC DATA

-------
100,000
0.02	0.04	0.06	0.08	0.10
NOMINAL FUEL AIR RATIO FAN
0.12
O.M Figure
VI-10

-------
10,000
TIQM VAPOR
1000.
100
a.
a.
w
Q
~H
X
O
as
o
o
3
3 10

A
a
¦- A
cr
. co
l.oj
0.0
0.01
0.02

0.03
0.
04
NOMINAL FUEL AIR
RATIO -
FAN

AIR FLOW
PROP.
KER.
FLAG
AIR
FUEL
#/Hr.



TEMP.
TEMP.
Under 40
+
+
A
Cold
Cold
40 - 70
O
•
A
Hot
Cold
70 -120
A
~
2v
Cold
Hot
120 -150
~
¦
2
Hot
Hot
Over 150
O
•



0.05
0.06
Figure
VI-H

-------
RUN
MO.
360
361
362
363
364
365
366
367
368
369
370
CO EMISSIONS COMPARISON
VAPOR GENERATOR VS BURNER
KEROSENE LEAN COMBUSTION
CO-top
PPM
10
14.6
RUNS WITH NO CO DETECTED WERE
PLOTTED AS 5 PPH WHICH IS THE
RESOLUTION LIMIT OF THE GAS
CHROMATOGRAPH

-------
i k
•if
PAXVE, BljlRNER EMISSIONS
HYDROCARBONS, ^ROM Tr E
BURNER !
1	VS i I
FUEL !AIR RATIO i
4
A
*
A
A

o o
O o
A
&
A
£
&
\
Vq a
A*i<*
Ef
H Gf Ef
^f'rM Mt
—13—
&
RUMS WITH ZERO HYDROCARBON
READINGS ARE PLOTTED BELOW
THE SCALE ON SEMILOG PLOTS
0.02
0.04
Air Flow
# / Hr
Prop
Ker
Flag
Air
Temp
Fuel
Temp
Under 40
4-

A
cold
cold
40 - 70
O
•
A
hot
cold
70 -120
A
~
£
cold
hot
120 -150
~
¦
&
hot
hot
Over 150
O
•



0.06
0.08
0.10
0.12
o.lu
NOMINAL FUEL AIR RATIO - (FAN)
Figure
VI-13

-------
HC
ppm
c -*1
1-1 £•
II 30
-p 2
(V
0.0"?
0.03
0.04


<




>

T3




>


O
X



70
-<
<



o
n
s

O
F1
§
CO


z
o
c

<
M
>
§
n
>
HH
C/5
5
g
so

H
O
o
z
to
5

PO
C/)
n
3
H


^1
HH
hH

CO
P0
C/)
o

-J
o
cn


>
"g
n


o

O


7t

z
cn
0.05
NOMINAL FUEL AIR

-------
PAXVE BURNER EMISSIONS
NITROGEN OXIDES FROM BURNER
vs
FUEL AIR RATIO
I
'TTU LttiJftfllUlilMlllil
AIR FLOW
PROP.
KER.
FLAG
AIR
FUEL
#/Hr.



TEMP •
TEMP.
Under 40

+
A
Cold
Cold
40 - 70
O
•
A
Hot
Cold
70 -120
A
~
X"
Cold
Hot
120 -150
~
¦
2
Hot
Hot
Over 150
O
#



IE
11
0.02	0.04	0.06	0.08
NOMINAL FUEL AIR RATIO (FAN)
0.10
figure
VI.I5

-------
PAXVE BURNER EHISSIONS
NITROGEN OXIDES FROM BURNER
WITH VAPOR GENERATOR OPERATION
NOMINAL FUEL/AIR ''ATIO (TAN)
AIR FLOW
PROP.
KER.
FLAG
AIR
FUEL
#/Hr.



TEMP.
TEMP.
Under 40
+
+
A
cold
Cold
40 - 70
O
•
A
Hot
Cold
70 -120
A
~
X'
Cold
Hot
120 -150
~
¦
2
Hot
Hot
Over 150
O
•



Figure
VI-16
DAXVE BUR'TR EHISSIONS
NITROGEN OXIDES
TROti VA"OR GENERATOR STACK VS
NOMINAL FUEL AIR RATIO - FAN
AIR FLOW
PROP.
KER.
FLAG
AIR
FUEL
#/Hr.



TEMP.
TEMP.
Under 40
+
+
A
Cold
Cold
40 - 70
O
•
A
Hot
Cold
70 -120
A
~

Cold
Hot
120 -150
~
¦
A
Hot
Hot
Over 150
O
#



VI-I7

-------
PAXVE BURNER EMISSIONS
NITROGEN OXIDES COMPARISONS
	-t-
W ^
~hT
<
\>
ifii
Hit
31
ti
gj
itf
A
I I
AIR FLOW
PROP.
KER.
FLAG
AIR
FUEL
#/Hr.



TEMP.
TEMP.
Under 40
+
+
A
Cold
Cold
40 - 70
O
•
A
Hot
Cold
70 -120
A
~
A
Cold
Hot
120 -ISO
~
¦
A
Hot
Hot
over 150
O
•



1.5
¦»	6 8 10	2
NITROGEN OXIDES TROH BURNER - PPM
6 7 B 9 ID
Figure
VI-18
&
e
S3
£
3600.0- +
3'too.o- ¦+
3200.0- +
3000.0- +
2600.0- +
2600.0- +
2400.0- +
2200.0- +
2000.0- +
1800.0-
TBURN F
t-
0.01	0.10	1.00 10.00 100.00 1000.00
OXIDES OF NITROGEN EMISSIONS FROM THE BURNER — NOB—-PPM
CORRELATIONS OF OXIDES OF NITROGEN DATA
Figure
VI-19

-------
PAXVE BURNER STABILITY DATA
Propane Ambient
FUEL AIR RATIO - f/a
°*3	0.4	0,5	0,6	0,7	0.8
E1UIVALENCE RATIO - (p
Figure
VI-20
PAXVE BURNTR STABILITY DATA
Propane Ambient
Fuel Air °atio - f/a
0.03	O.U	0.5	n.6
valence n?tic
rigure
'1-21

-------
200
PAXVE BURNER STABILITY DATA
PROPANE HOT
TOEOJOTXCAL-
STABILITY
LI HIT «*oo F
O " Stabl*
A- Llait
~ - Goo Out
Burn»r Vol.ua* • 52.3 in3
<&© O O
G
O

-------
PAXVE BURNER STA3ILITV DATA
Kerosene - Ambient
Fuel Air Patio - f/a
I	1	1	3	1	
0.3	0.4	0.5	0.6	O.t
Eouivalence Patio - (t)	' r
T	2^
•*AXVr VJPfE^ STABILITY DATA
Kerosene - Hot
o.o>	o.nii	r.05
n-el *:r Datio -
	1			
r"u_vnienc» - cTt.»"»

-------
SUMMARY OF PAXVE BURNER STABILITY DATA
200
150
100
50
d
cr
l^<08^a' 0 cC
6 O
 O QD
O Stable
A Limit
~ Goes Out
Open Symbols - ^ropane
filled " - Kerosene
Flag (cO Indicates Hot
Fuel and Hot Air
	1	i	i	i	1	1	i	1	i	
0.02 0 .04 0.06 0 .08 0 .10 0.12 0 ,1U 0.16 0 .18 0 .20
Fuel Air Ratio f/a
¦i	1-
0.4 0.6 0,0 1.0 1.2 1.4 1.6 1.8 2.0
Equivalence Ratio - (£)
2.5
3.0
Tig VI
26

-------
ion,ooo
10,000
1000
100
0.05	o.06
NOMINAL FUEL AIR RATIO-FAN

-------
100,000
10,000
Q 08	0 .10	0 .iz
NOMINAL FUEL AIR RATIO -FAN
0 .16
Figure
VI-28

-------
100,000
10,000
1,000
0'Ql*	o .05	0 .0C
PUTL AIR RATIO-FAN
0 .07	0 .OF,
Figure
VI-29
10,000
1000
100
i+T
~n—

TO5
tt:
T
-4-4-
L t	r t'
-------
10,000
r-TlT
111 r 11:1. i1111111111—1—i	1	T7—r
CO EMISSIONS DATA FROM THE PAXVE BURNEP
FUEL :PROPANE
AIR TEMP sOVER 250 F
BURNER VOLUME : 52.3 CU IN
A
o
S3'
10
I
: "~© '
•*I
AIR FLOW
#/Hr.
Under 40
40 - 70
70 -120
120 -150
Over 150
+
O
A
~
O
Flag indicates runs
from No. 282 ON
1"—
0 .02
EQUILIBRIUM CO
RUNS WITH NO 00 DETECTED WERE
PLOTTED AS 5 PPM WHICH IS THE
RESOLUTION LIMIT Or THE GAS
CHR0MATOGRAPH
0 .03	0 .0M	0.05
t'0'IIVAL rUHL Air
0 .06
0 .07
Figure
VI-31
CO EMISSIONS DATA FPC' THE P/J"L UUP-.I'
FUEL :PPCPmi
AIR TH.-P :OVTP 250 T
BURNER VOLUME : 52.3 CU IN
VAPOR GENERA TOP EXI'ACST DfTf
10,000
Figure
VI-32

-------
100,000


CO EMISSIONS D/TA FPOI Tlir FAXVE BUR-IE"
FUEL : rFCIYHI
HP 1Et!T	250 F
BURNER VOLUME : 52.3 CU IN
wpof crurr/iTon r:n /lst
n, ooo
1 or1"
AIR FLOW
#/Hr.

Under 40
+
-
40 - 70
O

70 -120
A
-
120 -150
a

Over 150
o

Flag indicates runs
from No. 28 2 ON
cf:
-4- -

££
RUNS WITH NO CO DETECTED VEKE
PLOTTED AS 5 PPH WHICH IS THE
RESOLUTION LIHIT OF THE GAS
CHROHATOGRAPH
0.0^	0 .01	0 .0"	0 .1r	0 -"r
. „.T,,	-r „	FiRure
- - RATIO —FAH VI-33
10"!0
100
n
0.T1
0."^
o.o"
O.O^	O.cr
" RATIO-FAH
0.07
Figure
VI-34

-------
10,000
1000
100
0.02
RUNS WITH HO CO DETECTED WERE
PLOTTED AS 5 PPM WHICH IS THE
RESOLUTION LIMIT OF THE GAS
CHR0MATOGRAPH
±±2
O.OU	0.05	0,06
'lOriNAL FUEL AIP SAT 10 -TAN
0.07
Figure
VI-35
0 .03
0 .05
0 .OC
0 .07
'I *«L ri^L ATP RATIO-FAN
Figure
VI-36

-------
,000
1000
0.02	0 .03	0 .on	0.Oc
t.n 'T'.AL FIFL AI"
O.Of)
0.07
Figure
VI-37
1000
100
10
> 1.0
o
<_>
.10
0.02
0.03
0 . 0U
0.07
FUEL TIP RATIO-FAN
Figure
VI-38

-------
0.02	o.03	0.04	0.05	0.06	0.07
NOMINAL FUEL AIR RATIO - FAN
0.08
0.09
figure
VI-39

-------
1000
0.03
0.04
0.05
0.06
0.07
0.08
NOMINAL FUEL AIR RATIO -FAN
COO EMISSIONS DATA FROM TIIE PAXVE BURNER
FUEL :PROP7*Nr
AIR TEMP :OVER 250 F
BURNER VOLUME : 33.0 cu IN
Figure
VI-40

-------
100,000
10,000
1000
0.02
0 .04
0 .06
a 08	0 .10	0 .12
NOMINAL FUEL AIR RATIO-FAN
figure
VI-i»i

-------
COG EMISSIONS DAT/1 FPOM Tlir PAX^ BURNEP
FUEL : PPOPANF
AIP TEMP : OVFF 250 F
BURNER VOLUME : 52.3 CU IN
m
TT
~rT
Tti
[fi
Si
3
i-n

TT
TT
t ?:
M J*

¦ I I i
I HI
1
: ' ; :
' 'L


n i i
11 i F
tifcf

rt>
ft"
83-
m
. 4
:: : i
x
Hit
AIR FLOW
l/Hr.
fill
Under 40
+
40 - 70
o
70 -120
A
120 -150
~
Over 150
O
Flag indicates runs
S~
w
0.02	0.03	O.on	0.05	0.06	0.07
NOMINAL FUEL AIR RATIO - FAN
Figure
VI-K2
1000
100
10
eo
s
1.0
0.10
0.02	0.03	0.014	0.05	0.06	0.07
NOMINAL FUEL AIR RATIO
COG EMISSIONS DATA FRO'" THE PAXVT BURNEP
FUEL : PPOP/I.r
Air TEHP :UNDrP 250 F
BURNEP VOLUME : 52.3 CU IN
VAPOR GENEPATOR EXHAUST DATA
AIR FLOW
#/Hr.
Under 40
+
40 - 70
O
70 -120
A
120 -150
~
Over 150
O
Flag indicates runs
from No. 282 ON
RUNS WITH NO CO DETECTED WERE
PLOTTED AS 5 PPH WHICH IS THE
RESOLUTION LIMIT OF THE GAS
CHROKATOGRAPH
Figure
VI-4 3

-------
1000
100
COG EMISSIONS DATA FROM Tlir PAXVE BURNER"
FUEL sPROPPNr
AIR TEMP :OVER 250 F
BURNER VOLUME : 52.3 CU IN
VAPOR GENERATOR EXHAUST DATA
10
1.0
.10
0.02	0.03	0.04	0.05	0.06
NOMINAL FUEL AIR RATIO-FAN
0.07
Figure
VI-44

-------
a02	o.03	0 .0«	o.05	o -06 0 -07
NOMINAL FUEL MR RATIO — FAN
0•08	o • 05
figure
VI-ns

-------
.10
0.02	0*03	0.0U	0.05
NOMINAL FUEL AIR RATIO — FAN
0.06
0.07
Figure
VI-46
1000
10
1.0
0.10
0.02	0.03	0.0<4	0.05
NOMINAL FUEL AIR RATIO-FAN
0.06
0.07
Ficure
VI-U7

-------
100
0.06
0.07
NOMINAL FUEL AIR RATIO - FAN
Figure
VI-U8
IiC EMISSIONS DAT/ FROy THE PAXVE BURNFF
FUFI. : PROPAIir
AIP TEI'P : UNbE" 250 F
BURNER VOLUME : 52.3 CU IN
VAPOR GENERATOR EXHAUST DATA
RUNS FROM 282 ON
0 .02
0 .03
0.0U	0.05	0.06
iCMT'iAL rUrL AIP RATIO - FAN
007
Figure
VI-U9

-------
NOMINAL FUEL AIR RATIO - FAN
Figure
VI-SO
HC EMISSIONS DAT* FPCM TKP PfJCVT BUPNI R
FUEL :PROP*NE
AIP TEMP sOVTP 250 T
BURNER VOLUflE : 52.3 CU IN
VAPOR GENERATOR LXU^UC? DATA
RUNS FROM N? 282 ON
i.O
0 .10
0 .01
ao:?
0.03
0 .04	0 .05	0 .06
"OMp^L nri \l° RATIO — FAN
0 .07
VI

-------
HC EMISSIONS DATA FPOM THE PAXVE BURI1EP
FUEL :KEPOPENE
AIR TEMP rUNDEP 250 F
BURNER VOLUME : 52.3 CU IN
BUKNER DATA
RUNS AFTER N? 282
1000 -
-J-l ,
cr
a
100
10 --
1.0 -
air flow
t/Hr.
Under 40
40 - 70
70 -120
120 -150
Over 150
+
O
A
~
O
Flag indicates runs
from No. 282 ON
££
a
o.i
0.(52
oTiJj
0.04
	1—
0.05
0 .06
0 .07
0 .08
O.i
NOMINAL FUEL AIR RATIO — FAN
VI
Fij. 52
0 .02
0 .03
0 . O'i	0 . or	0 , 06
• uf.l r:~ ratio -fan

-------
HC EMISSIONS DATA FROM THE PAXVE BURNER
PUEL (KEROSENE
AIR TEMP iUNDER 230 P
BURNER VOLUME : 52.3 CU IN
VAPOR GENERATOR EXHAUST DATA
RUNS AFTER NV 282
0 .01
0.02
0.03
0.0U
0.05
0.06
0.07
NOMINAL FUEL AIR RATIO-FAN
VI
Fir,. 54
HC EMISSIONS DAT/1 FFO:1 THE PAXVE Dl'PTJV P
rur-L sKrposrwr
AIF TEI'iP :OVLT 250 F
BURNER VOLUn: : 52.3 CU IN
bOOr DATA
rums Arr° 2n2
1.0
o.io
0.01
0 .06
O .07
NOMINAL FUEL AIR RATIO — FAN
VI
1 1 .DO

-------
0.02
0.03
0.04	0.05	0,06	0.07
NOHI'IAL niKL AIR RATIO - FAN
0.08
0.09
0 .01
0 .02
0.0't	0.0 5	O-01"'	0*n7
•|0;iI\AL fLTI. AIP RATIO — FAN
0 .08
0.0?
VI
Fig.56

-------
1000
100
1.0
0.1
0 .05	o .06	0 .07
NOMINAL FIIFT. ATR RATIO — FAN
0 .08	0 .09
VI
Fig.58

-------
0.02
0.03
0 .04
0.05
0.06
0.07
NOI'IHAL FUEL ^IF RATIO-FAN
VI
Fig.59
HCG EI'.ISEIOIIF DAT/ FPOV THE PAXVE BUR1JFP
fuel :rpop;un
flP TEH :l!NDFr 250 T
BUPNEP VOLUriF : 52.3 CU IK
VAPOR rXNr^ATOR EXIIALST DAT?
RUNS FROM 2B2 ON
10
0. 01
1.0
0>
M
D1
= 0.1
0 . 02
.03
°.04	0.05	0.06
IJOMIMAL FULL AI p RATIO — FAN
0 .07
VI_.
71 . Ca

-------
100
0 .0)
0 02	0.03	0.04	0.')b	0.'16	0.O7
NOMINAL FUEL AIR RATIO-FAN
0. 3',
0. V
VI
rip.6i
1.0
0.1
0.01
0.03	0.
NOMINAL nT
04	0.05	0.06
'I MS RATIO — PAN

-------
0,03'
0 .04
0 .05
0.06
0 .07
NOMINAL FUEL AIR,RATIO - FAN
VI
Fig.63
-in
HCG EMISSIONS DATA FROM THE PAXVE BURNFP
FUEL sKEPOSFNE
AIR TEI1P :OVTR 250 F
mi
o.oi -

NOMINAL FUEL AIR RATIO - FAN
VI

-------
HOtllNAL rUKL AIR RATIO — FAN
VI
Fie,. 65

-------
1000
0.0M	0.05	0.06	0.07
NOMINAL FUEL AIR RATIO -FAN
0.09
VI
Fig.66

-------
.05	.06
NOMINAL FUEL AIR
VI
Fig.67

-------
0.02	0.03	0.04	o.05	0.06	0.07
NOMINAL FUEL All RATIO - FAN
0.08 yj
Fig.68

-------
1000
0.02
0.03
Q.OH	0«05	0.06	0.07
NOMINAL FUEL AIR RATIO - FAN
0.08
0,09


-------
NOx EMISSIONS DATA FROM THE PAXVE BURNER
FUEL
AIR TEMP
BURNER VOLUME
PROPANE
UNDER 250°F
66.5 CU IN
10r-
O.Ol
0.001
0.02
0.03
0.03
0.05
0.06
0.07
0.08
0.09
NOMINAL FUEL AIR RATIO - FAN
r-. VI
Fi
-------
0.001
0.02
0.03
°iW	0.05	0.06
NOMINAL FUEL AIR RATIO - FAN
0.07
VI
Fig.72
0 .001
0 .02
0 .03
0 «0U	o
NOHIANAL
05	(>06	0 .07
FUEL AIR RATIO - FAN
0 .08
0 .09
VI
"I .73

-------
0.001
0.04	0.05	0.06
NOMINAL FUEL AIR RATIO-FAN
VI
Fig.74

-------
10,000 -r
1000
100
Z
a
a
S
X
o
z
1.0
0.1
0.02
0.03
0.04	0.05	0.06	0.0".
NOMINAL FUEL AIR RATIO — FAN
0.08
0.09
yi
rig.75

-------
10,000
9.02
0 .03	o -04	0.05	0.06	0.07
NOMINAL FUEL AIR RATIO - FAN
0.08
0.09
VI
Fig.76

-------
io,ooa
0 .02	0.03	0 .04	0.05	U06	0.07
NOMINAL FUEL AIR RATIO-FAN
.08
VI
fig.77
.09

-------
10,000
0.02	0.03	0.04	0.05	0.06	0.07
NOMINAL FUEL AIR RATIO-FAN
0.08
0.0'
VI
Tig.78

-------
NOX EMISSIONS DATA FROM THE PAXVE BURNER
FUEL :PROPANE
AIR TEMP iUNDER 250 F
BURNER VOLUME : 52.3 CU IN
VAPOR GENERATOR EXHAUST DATA
1.0
J?
a 01
0.02
0.03	O.OU	0.05	0.06
NOMINAL FUEL AIR RATIO - TAN
0.07
rig.79
NOX EMISSIONS DATA FROM THE PAXVT BURN I P
FUEL : PFOPAN]
AIP TLTP :OV£R 250 F
BURNER VOLU*IE s 52.3 CU IN
VAPOR GENERATOR ZXriAUST DW
0.001
0*02	0.03	0.04	0.05	0.06
NOMINAL FUEL AIR RATIO - FAN
0.07

-------
s o.oi
0.001
0.02
0.03	0.0M	0.05	0.06
NOMINAL FUEL AIR RATIO -FAN
0.07
Fig.81
1.0
0.1
"0. 01
0 .001
o .oe
0 .07
NOMINAL FUEL AIR RATIO — FAH
ri .c;'

-------
VII. ANALYTICAL INVESTIGATION
During the course of the program Paxve conducted several
analytical studies related to burner operation. The purpose of
these studies was to model the performance of a burner to determine
which parameters affect the stability limits, the completeness of
combustion, and the burner emissions. It was found that a
simplified well stirred reactor model could be used to correlate
some of the experimental data. The background for that analytical
study is presented in Section A below. The analysis itself is
presented in Sections B and C.
A. Literature Survey
Combustion theory has been the subject of a wide
variety of analytical and experimental investigations for the
last several decades. Analyses have included ignition theory,
flame stability theory for flame holders and enclosed burners,
and analysis of operation and stabilization in recirculating
burners.
1. Ignition Theory
Quoting Frank-Kamenetski (Reference 1), "The
basic idea of the theory of thermal ignition is due to Van't
Hoff (Reference 2). According to it, a condition of thermal
ignition consists in the impossibility of a thermal equilibrium
between the reacting system and the surrounding medium. The
qualitative formulation of this condition as contact between
the curve of heat supply and the straight line of heat removal
was first given by Le Chatelier (Reference 3). The mathematical
formulation was given by Semenov (Reference 4) who obtained an
expression for the relation between the explosion parameters
(temperature and pressure at the explosion limit) which was later
confirmed by Zagulin and a number of other investigators."
The essence of the thermal ignition theory is illustrated
in Figure 1 which shows the heat release and heat loss from a vessel
in which a combustible mixture has been placed. The walls of the
vessel are heated to a temperature Tw and the combustible material
inside the vessel undergoes chemical reaction governed by an
Arrhenius type equation of the form
qj = a e"EA/RT
where
Qj = rate of heat release in the gas
B = a constant which takes into account the concetration
of the fuel and oxidizer within the vessel and other parameters
significant to the rate of the chemical reaction.

-------
Ea = the activation energy of the combustion reaction
T = the gas temperature
R = the universal gas constant
The gas mixture in the vessel loses heat to the walls
in accordance with an equation such as
Qll = hA(T-Tw)
where
Qll = rate of heat loss from the gas to the wall
h = heat transfer coefficient from gas to wall
A = wall surface area
Tw = wall temperature
Semenov examined these two curves, Qj the heat
production curve^ ancjl Qjj the heat loss curve. When the
two curves intersect (line 3 in figure 1) the gas exists at
a temperature slightly higher than the temperature of the wall
(Tw3> with heat production in the gas being carried to the wall
as a result of the minor temperature difference. For a higher
wall temperature, Tw,, the two curves are tangent, and (case 2)
the temperature of tne gas rises, but there is still a point,
Ta, at which, the rate of heat production and the rate of heat
loss to the wall are equal. Hence a stable system can exist.
At still higher wall temperature, , there is no point
of intersection (case 1). The rate of heat production in the gas is
now always greater than the rate in which heat can be lost to the
wall. Thus in case 1, if the gas starts out initially at the
wall temperature Twj, the temperature of the gas will increase and
although the rate of heat loss will also increase the heat
production will always exceed the heat loss rate. Once the point
Ta has been reached, the rate of heat production will accelerate
and the temperature of the gas will increase at an exponential
rate.
Semenov termed the value of T^/ f°r which the tangent
point case occurs, the adiabatic explosion temperature. The
adiabatic explosion temperature depends not only on the factors
influencing rate of heat liberation, but also on the surface
area of the vessel. The time required for the gas to reach the
exponential temperature rise situation is called the induction
period. Semenov examined these factors mathematically in some
detail.
2. Stability Theory for Combustion Chambers
Vulis (Ref. 5) performed an extensive analysis
of the problems of furnance combustion using Semonov's thermal
VII-2

-------
ignition theory as a starting point. Vulis argued that, just as
the rate of heat production and the rate of heat loss could be
used to compute the ignition temperature, one could also compare
the rate of heat production with the heat required in a flowing system
to analyze the combustion process. The heat required is that needed
to heat the inflowing gases to the chamber condition and to make
up heat losses from the burner. Vulis analysis is very extensive
and covers a great many of the cases of interest to us, and goes
beyond those cases to consider in some detail the influence of
heat loss by radiation and convection. He only considered first
order reactions, however, which limits but greatly simplified
the analysis, vulis also considered the flow case in which heat
is transferred back to the incoming gases by radiation and mixing.
Vulis' analysis for burner stability follows substantially
the same lines as the analysis presented in Section VII B of this
report. It differs from our analysis in two resoects. First,
Vulis considered only first order combustion kinetics. This greatly
simplifies the analysis, but at the same time it restricts the
validity of the results. In partiuclar, the influence of
pressure on stability tends to be lost with this approach. The
other difference between Vulis' analysis and that presented
below lies in the grouping of parameters used in the solution.
Vulis defined a considerable number of non-dimensional parameters
whose arrangement was convenient for his analytical work. Vulis1
parameters, however, tend to obscure the heat release and the
sensible heat requirements of the analysis. In our analytical
work we have used parameter groupings of a more conventional type.
Our analysis in this regard follows more closely the work of
Longwell and Weiss.
Vulis' analysis, in common with Semenov, assumes that
the rate of the combustion reaction is given by an Arrhenious
equation. Vulis writes
Qj = ko c q e E^/RT
where
Qj = heat release rate
ko = a constant
c = concentration of reacting material
q = heat release of the reaction
Ea, R, T = as before
This heat release is assumed to take place within a chamber as
shown in Figure 2. Here combustible material enters the chamber
from the left while burned material leaves the chamber on the right.
It is assumed in the analysis that mixing of the unreacted material
with the burning material in the chamber takes place instantaneously
and continuously with zero mixing time and distance. It is also
assumed that the concentration of reacting material varies
VII-3

-------
linearly with the temperature. In other words, as C goes from its
initial value CQ to zero, the temperature increases from the inlet
to the theoretical flame temperature.
The heat supply required to raise the incoming material
to the temperature at which it leaves the chamber can be expressed
in terms of the sensible heat of the reacting material
Qjl =	W Cp (T-T0)
where
Qn =	sensible heat increase in the exhaust products
W =	flow rate of reacting material into the chamber
Cp =	specific heat of the exhaust products
T = temperature in the chamber which is the same as
the exhaust temperature
T0 = inlet temperature
Figure 3 shows the heat release (Qj) and sensible heat
(Qn) curves as a function of the temperature in the combustion
chamber. These curves are similar to those drawn by Semenov in
his ignition theory analysis. There are two differences, however;
in Semenov's analysis, only the lower portion of the heat release
curve was considered. Vulis, on the other hand, considers the
entire heat release curve from below the inlet temperature up to
the theoretical flame temperature. The other difference is that
the straight line in Semenov's theory represented heat loss to
the wall. In Vulis' theory, the straight lines represent the
sensible heat of the exit flow.
A stable situation for both Vulis' and Semenov's analyses
is represented by an intersection of the two curves for which the
heat requirement curve, Qn, increases more rapidly to the right
of the intersection than the heat supply, Qj. An intersection for
which Qj increases faster than Qjj is an unstable point which
cannot correspond to a steady state solution. This matter is
discussed more fully under Section VII B below.
Vulis considered a number of concepts important in
burner operation which can be understood by reference to the balance
between the heat release rate and the sensible heat required.
Figure 4 shows the influence of changing the inlet temperature. The
effect here is very similar to that discussed by Semenov when he
discussed the influence of changing wall temperature. At some
sufficiently high inlet temperature, Tjj, there is only one
intersection and hence only one operating point for the system.
That intersection corresponds to stable combustion taking place
within the chamber. This inlet temperature represents and gives
rise to the situation in which spontaneous ignition occurs as
the gas flows into the chamber.
VII-4

-------
If we lower the inlet temperature to the value labeled
1l2i spontaneous ignition no longer takes place. If there is no
combustion occuring in the chamber, the gas will flow through the
chamber and the stable intersection to the left of the figure which
represents only a small temperature rise, will pertain. If on the
other hand, combustion has already been initiated, then the stable
intersection to the right will pertain, and the system will continue
to burn even though we no longer have a temperature above the
ignition temperature at the inlet.
If we further reduce the inlet temperature sufficiently, we
can in principle cause the burner to go out. The incipient
extinction condition corresponds to T13 in Figure 4. Here, there is
only a single point of tangency between the heat release curve
and the heat required curve that corresponds to combustion within
the chamber. A further reduction in inlet temperature will
eliminate this solution which permits combustion to continue, and
the burner will go out.
In a similar fashion, Vulis observed that air flow
could be varied in such a fashion as to allow ignition or
extinction of the chamber. Figure 5 illustrates the influence of
varying the flow rate. Here the lower straight line labeled Wj
corresponds to the flow through the chamber for which spontaneous
ignition will take place. The flow line W2 allows for stable
combustion to continue, provided it has somehow been initiated.
The flow rate corresponding to W2 will not, however allow the
burner to spontaneously ignite. A further increase in flow to
W3 corresponds to incipient blow-out. Any further increase in flow
will result in a situation in the burner for which no stable
combustion solution exists, and the burner will be extinguished.
Another feature of combustion chambers which was noted
by Vulis, is that for a sufficiently high inlet temperature, the
critical phenomena of ignition and extinction do not occur. Such
a situation is illustrated in Figure 6. We see that for a
continuous variation flow rate from a very low value to a very
high value, no tangent points exist between the straight line
family of curves (Qji) and the heat releases curves (Qj). At
this inlet temperature, only combustion solutions are possible
since there is only one intersection between Qjj line and the Qj
curve. The combustion "becomes increasingly efficient as the flow
rate is reduced, but even at high flow rates some combustion takes
place.
Vulis considered not only variations in flow and inlet
temperature, but also variations in other significant parameters
such as the heat of reaction (q). Additionally, he considered the
influence of factors such as heat loss and flow recirculation on
performance of combustion systems. By restricting himself to
first order reaction kinetics he was able to handle the
mathematical details of his analysis and provide generalized curves
which are of great interest.
VII-5

-------
3. Gutter Burner and Can Burner Stability
Dezubay (Reference 6), Scurlock (Reference 7)
and others have conducted experiments on flame stabilization on bluff
bodies. They were able to show that the flame holding action of a
bluff body such as a disc or V shaped gutter can be correlated in
terms of the flow velocity by the flame holder, the dimensions of
the flame holder, the pressure level in the burner, and the fuel/air
mixture ratio. Dezubay's empirical correlations were of the form
shown in Figure 7. The ordinate in Figure 7 is the fuel/air
ratio at the flame holder. The abcissa in Figure 7 is a combination
of burner parameters given approximately by V/PD.
Longwell and Weiss (Reference 8) conducted
experiments similar to those Dezubay except that their flame
stabilization testing was done for a can type burner. Burner
stability correlations for can burners are generally similar
to those for gutter type burners. Figure 8 shows a typical can
burner flame stabilization curve. The ordinate again is fuel/air
ratio. The abcissa in this case is the burner intensity parameter
I = WA/Vol p2.
Longwell and Weiss also conducted analyses
similar to those of Vulis. They considered second order combustion
reactions. Longwell is generally considered the author of the
phrase "well stirred reactor theory" which covers the case
which Vulis called the "zero dimensional case" in which rapid
mixing of the inlet flow with the burning material in the chamber
is assumed to take place.
Because we will use "well stirred reactor"
theory in Section VII B, we will not deal extensively with the
details of this type of analysis here. It is interesting to note
however, that Dezubay's correlation parameter and the combustion
intensity parameter of Longwell and Weiss are closely related.
Figure 9 shows how each of these resembles the well stirred chamber
considered by Vulis. On the left hand side of Figure 9 we have a
sketch of the front portion of a can burner, the air flow W^ enters
first row of holes and circulates within the pilot region of the
burner, mixing with the burning gases which are contained therein.
The volume of the pilot zone and its pressure level are the other
parameters of significance in the intensity parameter. On the
right hand side of Figure 9, we see a sketch of the recirculation
region behind a gutter burner or disc type flame holder. As
illustrated here a separated region exists downstream of the
flame holder within which the material flowing past the obstruction
is recirculated and mixed with burning material which has been
stabilized on the baffle. The length of the separated
recirculation wake will be proportional to the characteristic
dimension of the flame holder, D. If we now consider the rate of
air flow into the recirculation volume we can write
Wa a PVD2
Similarly, the size of the recirculation volume should be given by
VII-6

-------
Vol a D3
Thus, if we form the ratio for the combustion intensity
parameter with regard to the separated wake on the baffle type
flame holder, we see that
Wa	„ PVD2 _ V
Vol p£ p7-DT PD
and therefore the two combustion parameter expressions are
equivalent.
4. Comments on Well Stirred Reactor Analysis
Analyses of the well stirred reactor concept have been
carried out by many investigators. References 9, 10, 11, and 12
as well as many others deal in various degrees of sophistication
and elaboration on the concepts set forth in the works of Vulis
and Longwell. It is interesting that this type of analysis is
applicable to virtually any type of chemical reaction process in
which heat release and a thermally controlled reaction rate are
significant. Zwick and Bjerklie conducted analyses on the
thermal decomposition of a monopropellant in a gas generator. Their
analyses included a well stirred reactor approach as well as one
dimensional kinetics approach. The well stirred reactor type of
analysis was shown to give a means of correlating experimental data
which agreed very well with the experimentally observed behavior of
the monopropellant gas generator.
Well stirred reactor theory is quite useful in that it
provides a means of correlating experimental data and predicting
the influence of various parameters on burner operation. It is not
in general an accurate description, however, of the actual situation
existing within the chamber. In a well stirred reactor analysis,
we assume that the temperature, pressure and compositon everywhere
within the region being analyzed are uniform. In practice, of
course this is not, and in fact cannot be true. It is not
surprising therefore that real chambers show deviations from
the predictions made by well stirred reactor analysis and that
the variations depend on the extent to which the processes in
the chamber lead to inhomogeneity and non-uniformity.
Longwell and Weiss fabricated a reactor which was designed
to be as close a physcial embodiment of thorough mixing as they
could acheive. Their analysis proved capable of correlating their
experimental data quite closely, which is not too surprising since
they had attempted to physically simulate the mathematical model.
More recently however, Reference 13, has shown that even in an
experimental well stirred reactor comparable to that used by
Longwell and Weiss, details such .as the size and location of the
injection ports and the magnitude of the injection velocity
influence the behavior of the system. This is of course what one
might expect as a result of non-uniformity within the chamber
itself.
We must expect therefore that well stirred reactor theory
VII-7

-------
will provide insight, but not complete detailed information on
burner performance. In this regard it was inevitable that the
simple "well stirred reactor" theory developed by Vulis and Longwell
would lead to more elaborate treatments of burner behavior based
on the same basic concepts. Several of the references cited
above attempt to refine the analysis by examining in more detail
the internal flow pattern, the chemical reactions, or the
character of the combusiton products leaving the burner under
conditons of incomplete combustion. While the merit of these
refinements can be argued, it was our purpose in the present
program to use this analytical method as means of understanding
and interpreting the experimental behavior of a real burner.
With this aim in mind, we decided to use the simplest model and
analytical method which would involve the parameters of significance
in our experimental program. The analysis itself is presented
later in this section of the report. The numerical constants
required .for _the chemical reaction rate expressions were
obtained from Reference 14 which gives a review of the various
analytical procedures devised by other investigators and also
presents equations and constants which give the best fit
to available experimental data.
5. Recirculating Flame Stabilization Analysis
In addition to well stirred reactor theory,
there is another simple model of burner performance which also
provides some interesting insight into burner operation and
yet is relatively simple in basic concept. In the analysis by
Zwick and Bjerklie (Reference 9) this type of process was defined
as recirculation theory.
Vulis distinguishes between the well stirred reactor and
the recirculation cases by identifying one of them as the zero
dimensional case and the other as the one dimensional case.
The mode of analysis is illustrated in Figure 10.* Here
we assume instantaneous mixing of the recirculated portion of the
exhaust products with the incoming stream of combustibles.
After the mixing takes place, we follow the combustion process
in the mixture as a funcion of time. The material which finally
emerges from the chamber differs from the material present at
the initial mixing point because of the reaction which takes place
during passage through the chamber. This material now represents
both the effluent from the chamber, and input to the recirculation
pattern.
Recirculation theory leads to predictions of chamber
performance which are similar to those of well stirred reactor
theory. They include, however, an additional parameter, the
degree of recirculation. When the degree of recirculation
approaches infinity the two analytical procedures yield the
same result.
Recirculation theory has an advantage over well stirred
reactor theory in that it allows one to consider a wider variety
of cases. Staged combustion which is common in many types of
VII-8

-------
gas turbine engine burners and two stage and multi-stage
combustion industrial and public utility boilers are burners for
which, recirculation theory provides additional insight while still
allowing for a simple and straight forward analytical procedure.
The true picture of what goes on within a combustion
chamber is of course quite complex. Multi-dimensional analysis
involving both space and time are required for an accurate model
of any real system. Unfortunately such analytical procedures
are extremely complex and further handicapped by the fact that the
flow patterns and mixing patterns which actually exist in a real
apparatus are sometimes unknown and almost beyond reach of any
reasonable analysis.
B. Burner Analysis
1. General
For the work conducted here it was decided to
perform a well stirred reactor type of analysis rather than to
engage m a more sophisticated recirculation type of study. The
Paxve burner has considerable internal mixing and hence should be
fairly well modeled by well stirred reactor theory.
The purpose of this analysis was two fold. First, we
wished to investigate the influence of various parameters such as
fuel/air (mixture) ratio and inlet temperature on the stability
of the burner. Early observations led us to believe that
the Paxve burner was stable over a wider range of operating
conditions than other burners with which we were familiar. The
possibility of exploring this analytically was therefore desired.
Secondly, we hoped that the analysis would shed some light on the
relationship between burner operating conditons and the
production of air pollution type emissions. In particular, the
degree of completness of reaction was to be determined to see
if this concept could serve as a means of correlating the
experimental data.
The model for the burner analysis is illustrated in Figure
11. Here the burner is represented by a chamber into which the
combustible mixture flows and from which the combustion gases
exhaust. Within the burner a homogeneous mixture is undergoing
chemical reaction. The rate of that reaction is assumed to be
given by an Arrhenius type equation. Because we were principally
interested in lean combustion we restricted our analyses to
mixtures which were leaner than stoichiometric.
The heat which is being generated by chemical reaction
within the chamber serves three purposes. First it raises the
incoming gas to the temperature within the reaction chamber.
Secondly, it sustains the reaction at a rate which is dependent
on that temperature. Thirdly, it supplies the heat which leaves
the chamber both in the form of hot gases in the exhaust and also
in the form of heat loss to the surroundings.
Chamber heat loss is potentially an important parameter
VII-9

-------
in the analysis. If the walls of the chamber are radiating to
the outside there will be heat loss which must be supplied to
those walls by convection and radiation from the combustion
gases. The chamber can also lose heat to the outside by
conduction through the chamber walls. This heat loss must be
made up by extraction of heat from the combustion process. Because
the stability and efficiency of the burner are very sensitive to the
combustion temperature, any heat loss will be significant. The
Paxve burner is a relatively well insulated chamber. To a
first approximation, therefore, we have ignored heat loss from
the burner.
2. Basic Flame Stabilization Analysis
The fundamental concepts involved in combustion
within an enclosed space such as the burner of a gas turbine engine
can be understood by reference to Figure 11. Here we see a volumne
with a fuel air mixture entering from the left at a given set 'of
initial conditions and combustion products leaving on the
right. The simplest method of analysis for such a system involves
the so called well-stirred reactor concept. The idea is that as the
material on the left enters the reactor, it mixes instantaneously
and uniformly with the material contained in the volume where the
combustion process is taking place. The material which leaves the
combustion chamber is assumed to have exactly the same properties
as the material contained within the chamber, and the exit flow
rate is assumed to be equal to the inlet flow rate.
The analytical method involves equating the rate of
heat release in the chamber, determined by chemical kinetic
considerations, to the rate of heat release required to raise the
gaseous exhaust products to their final temperature. In performing
this heat balance we make use of a concept which runs through the
entire analytical scheme; that one can consider the reaction as only
being partially complete. We use the Greek letter e to symoblize
the "reactedness" of the material passing through the combustion
chamber. It corresponds roughly to a combustion efficiency. For
a reactedness of 1.0, all of the material which enters the chamber
from the left leaves in the form of theoretical combustion products
and temperature in the chamber is the theoretical flame temperature
diminished by any heat transfer which occurs from the combustion
gases to the outside through the chamber walls. If e, is less
than 1.0, a portion of the exhaust products is unburned material
and the combustion temperature is correspondingly less.
The analysis is subject to various degrees of sophistica-
tion depending on how one treats the unreacted or partially reacted
material which leaves a chamber. The simpliest treatment, and
the one which will be followed here, is to assume that it leaves
as vaporized fuel. A more sophisticated approach which more nearly
fits the facts for lean mixtures of hydrocarbon fuels with air,
is to assume that the unreacted material leaves as the partial
reaction products which are water and carbon monoxide. The degree
of unreactedness is then represented by the failure of the carbon
monoxide formed to convert to carbon dioxide before leaving the
VII-10

-------
chamber. While this more sophisticated approach is in closer
agreement with reality, it introduces some difficulties in the
analysis which do not help to clarify the points which we wish to
make in this report. We will adopt the simpler approach here.
3. Mass Balance
Entering the burner are a flow of air, W^, and a
flow of fuel, Wf. Flowing out is a mixture of combustion products,
Wg, together with the unburned material, Wy. The chemical reaction
involved (at stoichiometric mixture) is
A + F ->¦ B + AHb
The unburned material exists in a state which may be different
from the original, perhaps involving only vaporization of liquid,
or possibly involving partial reaction.
Let
0 =	stoichiometric f/a
 =	(f/a) - 6
e =	fraction reacted
Then for lean	operation, the air which can burn with WF is
wab =
leaving
# • •
wAx = wa - p.
which is excess air.
The fraction of the combustible mixture which burns pro-
duces exhaust flow
WB = £ (wF + W&)
The excess air fraction together with the unburned portion
of the combustible mixture yields
WU = (wA - Je) + (wf +
If we divide the above expressions by the air flow, W^, and sub-
stitute and simplify we obtain
VII-11

-------
*? = E*<1+e>
Jg = (1-*) + *(1+0)(1-e)
The unburned material may be thought of as unburned air, WAu, plus
unburned fuel, Wp^,
then
WAu = WA(1-*) + WA4»{l-c)
= WA (l-e4»)
and
• •
wFy = Wp(l-e)
= WA <1)6 (1-e)
4.	Heat Balance
The heat release will be assumed to be given by
Ql = WBAHB
= WAAHBe (1+0) CpB + (l-e) CpA + (1-e) 0Cpp) AT + (1-e )AHy	[	
B T1 CpB ei(i(l+0) + CpA (1-e 0) + Cpp (1-e) G
5.	Reaction Rate
The reaction rate for the combustion reaction in
the volume is given by
WB = K [F]x [a]n_y	Vol e"E/RTB
VII-12

-------
where
K = K (T)
= collision factor

[F] , [a]
= concentration of reactants in mole fraction
X
= fractional order
= 1
n
= order of reaction
= 2
vol
= volume of chamber

E
= activation energy
of the combustion reaction
We can easily show that
[F ] =
and
	^	Wf/Mf	
Wb/MB + WAv/MA + Wp/Mp
(1-0)4)6		
(l-e)0 + (l-e<|>) Mf + e) Ma	
ldJ (1-e)<(>0 + (1— e) Mf + e Ma
'*<&L
vol e
-E/RT.
-e)d)e + (l-£) Mp + e<^(H8) Mp~
ma	mb
6. Stability Requirement
Stable operation will take place if the rate of
heat release is equal to the heat production necessary to heat the
exhaust products. There are several ways to approach this last
equation. Perhaps the clearest is the one which involves plotting
the heat release rate and the heat production rate required
against temperature in the burner. The result is two curves of
the sort shown in Figure 12.
In Figure 12 there are two lines, one labeled Qjj and the
other Qj. Qj is the rate of heat release based on the reaction
VII-13

-------
rate equation while Qjj is the heat flux necessary to achieve
the indicated temperature. In general these curves will have 3
points of intersection, one close to the origin and the other two
as noted in Figure 12 and marked a and b.
The intersection at point b represents a stable operating
point for the burner. For temperatures higher than T^, the heat flux
necessary to achieve the indicated temperature is higher than the
rate of heat generation available in the burner. As a result,
if the burner is initially at a temperature slightly higher
than T^, it will come back to the operating point indicated by b.
On the other hand, if the temperature is slightly less than the
temperature corresponding to intersection b, the rate of heat flux
in the system is greater than that necessary to heat the products
to the indicated temperature and hence the temperature in the
combustion volume will increase until the stable operating
point b is reached.
Point a is an unstable point. If the temperature is
slightly less than that indicated by the intersection at a, the rate
of heat production is less than the heat necessary to achieve the
indicated operating temperature. The temperature of the combustion
gases will fall towards zero and the burner will go out. If the
termperature of the gas in the burner is slightly higher than
Ta, the rate of heat production will exceed the rate of heat
required for the temperature in question and the temperature
of the gases in the burner will increase until stable point b is
again reached.
It is interesting that in practical combustion devices
the phenomena discussed above can be observed, sometimes with unusual
results. For example a small gas generator with which we have had
operating experience, could be inadvertantly placed into operation at
this partial reaction point for a matter of many minutes before it would
either jump up to point b and operate stable and efficiently, or else
quench and go out.
The heat release rate indicated by the curve Qj depends
on the volume of the system, its pressure and temperature, but
does not depend directly on the flow rate through the burner.
The curve labeled Qgai> on the other hand, is directly proportional
to the flow rate through the burner. If we increase the flow
rate, new points of intersection between the two curves will
be achieved. Figure 13 shows the limiting case where the two
curves are tangent at only one point which in this case is
labeled c. Point c is the incipient blowout limit of the burner.
The burner may operate at this point for an extended period of time,
but it has no stability margin. Any minor shift in conditions in
an unfavorable direction will cause the reaction to die away.
It should be noted that there are two ways in which
one can reach a point of incipient blowout in a given burner.
We can change the conditions of the line affecting Qn by
increasing the air flow, or we can change the factors affecting
Qj. The principal factor influencing Q React in an otherwise stable
situation is the equivalence ratio <)>. One might also, however,
VII-14

-------
change the reaction rate as a result in change of pressure. In
the experiments conducted by Paxve on the Paxve Burner for
APCO, we have determined incipient blowout by reducing the fuel/
air ratio until the lean limit was reached.
A factor which was not introduced in the above equation,
but which is significiant in a real burner, is heat transfer from
the gases. Heat transfer has the effect of reducing the heat
release indicated by the curve Qi by an amount which depends on
the temperature of the surroundings and the mode of heat transfer
involved. In particular, heat transfer to the wall of the burner
has a dramatic affect on the stability of the burner. A burner
of the type utilized by Paxve has very little heat transfer
to the walls under steady state operating conditions. During
startup and shutdown, and when changing operating conditions, heat
exchange between the gases and the wall plays a significant role.
As incipient blowout is reached, the burner will remain lit for a
matter of 10 to 20 minutes even though the equivalence ratio has
been reduced below blowout limit. The difference between the
heat required and the heat available in these conditions is quite
small, and the warm walls of the burner provide the necessary
difference during the time the burner takes to go out.
If we equate the heat production required to achieve a
given temperature with the heat production available as a function
of reaction rate, we obtain the following equation.
I= m= k(l-e)(l-efl) 6 e-EB/RTB	
Pz Vol e (1+0) [<(>(1+0) e + (l-e) M£ + (1-e) e4>]'TB 1. 5
?tb	Ma
The reaction rate constant K here has been related to temperature by
K = k T0-5
and x = 1 and n = 2 are assumed.
The equation for I can be solved if we have values for all
of the parameters and TB. The equation for TB requires that we have
values for the average specific heats of the air, the fuel and
combustion products. For the present analysis, equations
were generated for these specific heat values using data from
reference 15.
The specific heat data (in metric units) and the averaging
equations (in English units) were:
Air
cp = 6.557+1.477 x 10"3 T-2.148 x 10"6 T2 cal/Mole °K
CA,va = 0.22618 + 1.4147 x 10"5(T+T ) - 7.8229 x 10"10 (T3-T03)
9	/(T-T0) BTU/lb°R
VII-15

-------
Fuel
cp = 0.02 + 1.51 x 10"3T - 7.7 x 10"7T2 + 1.5 x 10"10T3 Cal/gm°K
CFavg = 0.02 + 4.194 x lO-* (T+To)-7.9218 x 10"9(T3-To3)/(T-To)
+ 6.43 x 10"12 (T^-To")/(T-To)	BTU/lb°R
CO?
c = 18.036 - 4.474 x 10"s T - 158.08 /\[¥	Cal/mole°K
CC02avg = 0.40991 - 2.8245 x 10"7(T+To)-9 .6403 (/F -fo)/(T-To) BTU/lb°R
H,0
cp = 6.970 + 3.464 x 10"3 T - 4.833 x 10~7 T2	Cal/mol°K
CH2°avg = 0-38722 + 5.345.7 x 10"5 (T+To)
- 2.7623 x lO-9 (T3-To3)/(T-To) BTU/lb°R
N2
c = 6.529 + 1.488 x 10"3 T - 2.271 x 10"7 T2	Cal/mole °K
CN? = 0. 23318 + 1. 4762 x 10~S(T+To) - 8.3444 x 10"10
^avg
(T3-To3 ) At-To) BTU/lb0R
For purposes of the analysis, the reaction taking place
was assumed to be that of octane with air. The reaction was given by
c8 h18 + |i (02 + 3.77 N2) + 8CO2 + 9H20 + 47.125 N2
This gives an exhaust gas composition of
N2 - 73.489 Mole% - 71.966 Wt%
C02 - 12.476 Mole% - 19.198 Wt%
H20 - 14.035 Mole% - 8.8356 Wt%
The average specific heat of the combustion products is therefore:
CBavg = 0.28072 + 1.5293 x 10~5(T+To) - 8.4458 x 10"10
(T3-To3)/(T-To)	BTU/lb°R
-1.8508 (\/T - ^To)/(T-To)
7. Limitations on the Analysis
The theoretical analysis performed here was
limited in two major aspects. First the influence of heat loss on
the stability and performance of the burner was not investigated.
Secondly, the performance and stability of the burner for fuel
rich operation were not investigated. The failure to investigate^
heat loss effects was the result of a lack of available
funds on the program rather than a lack of interest in the
VII-16

-------
subject. Failure to investigate fuel rich operation was due
primarily to a combination of lack of interest and the
factors sighted above.
The influence of heat loss on burner performance and
stability is a subject which should be studied closely. It may be
that the outstanding performance of the Paxve burner is in some
measure attributable to the low heat loss characteristics of this
device. Fuel rich operation is of some interest for burner
applications to systems in which two stage combustion would be a
desirable feature. Fuel rich operation of the Paxve burner is
not contemplated for automotive Rankine cycle or automotive gas
turbine applications.
C. Computer Analysis
The theoretical burner analysis equations presented
above were programmed for analysis on a digital computer.
The programs were written in APL, a new programming language
devised by IBM and made available on a time sharing basis
through Proprietary Computer Services, Inc., of Van Nuys. A
number of programs and sub-routines were written. The purpose
of each program and the results of the analyses are discussed below.
1.	Program CAL
Program CAL, shown in Table 1, is the basic
computation program used in the burner analysis. This program
was used as a subroutine in a number of the other programs. For
given input values of air inlet temperatures, (TI - degrees Rankine)
and equivalence ratio (PH), CAL computes the combustion temperature
(TB) and the combustion intensity parameter (INT).
CAL includes a correction to the heat of combustion which
allows for the heat required to warm the fuel up to the initial
temperature, and an iteration procedure for the specific heats
of the air/fuel and combustion products. Initial values for
these three specific heats are assumed in order to make an initial
estimate of the combustion temperature. The specific heats are then
corrected for the average value between the inlet temperature and
the combustion temperature and the process iterated until the
computed combustion temperature agrees with the previous value
to within one degree.
CAL returns values of TB (the combustion temperature), and
INT (the combustion intensity parameter) to the other programs.
2.	Program HOT
Program HOT,, shown in Table 2 was used to
compute tables of TB and INT for sets of inlet temperature,
equivalence ratio, and combustion efficiency values. HOT calls CAL
as a subroutine. HOT was used to generate the results presented
in Tables 3 through 8.
During the course of the contract, we frequently found
VII-17

-------
it desirable to be able to compare the predicted efficiency of the
burner with the experimentally measured values of carbon monoxide
and unburned hydrocarbons. In this regard, curves were made of
the so-called "unreactedness" which is defined as
6 = 1 - E
where
E = Burner Efficiency
Figures 14 thru 19 show the unreactedness plotted against the
intensity parameter I for various values of inlet temperature. The
parameter in the curves is the equivalence ratio PHI.
3.	Programs BURN and STABILITY
BURN is the computer program which was used to
compute the stability of the burner. BURN accepts an inlet
temperature TI as its input. It then examines the value of the
combustion intensity parameter, INT, over the range of burner
efficiencies from 0.3 up to 1.0. It uses the subroutine CAL
to perform this calculation. BURN then selects the maximum value of
INT, and calculates 10 new values, five on either side of the
maximum. It returns this value of INT together with the
corresponding value of burner temp. (TB) and combustion efficiency
(E) .
STABILITY is a small program shown in Table 9 which was
used to vary the imput temperature and accumulate the output for
the combustion stability calculations. It used BURN as a
subroutine for the stability calculation.
Tables 10 and 11 show the results of the stability
calculations made using STABILITY and BURN. The output columns
give the equivalence ratio, the efficiency at the stability limit,
the predicted combustion temperature at the stability limit
allowing for the inefficiency, and the combustion intensity
parameter at the stability limit.
Figure 20 shows a plot of the combustion stability data.
In this figure, the limiting value of the combustion intensity
parameter is the abscissa and the equivalence ratio is the
ordinate. The parameter for the curves is the burner air inlet
temperature.
4.	Program IHTPLOT
In performing the combustion stability analysis
with Program BURN, we assumed that the intensity parameter would
show a local maximum with varying efficiency. This is not
always the case, as was pointed out in the earlier discussion
of combustion theory. When the inlet temperature is raised to
a sufficiently high value we find that the critical conditions
which normally characterize burner operation are no longer
VI1-18

-------
present.
The normal phenomena in a burner involves both ignition
and extinction limits. For a given value of an inlet temperature
and equivalence ratio there are normally two air flow rates
(combustion intensity parameter conditions) which correspond to
these two critical values. At a sufficiently low value of air
flow rate, an adiabatic burner will spontaneously ignite. At
some higher value of air flow rate the burner is unable to
sustain combustion and the burner blows out.
This form of ignition only occurs in practice at high
inlet temperatures. It is not the one which is used in most
conventional burners. It is nevertheless a concept which is readily
apparent from the combustion analysis. High flow rate flame-out
is, however, a common occurence in any burner system.
In order to examine this question further, the values
of combustion intensity parameter corresponding to various
values of combustion efficiency and equivalence ratio were computed
using computer program INTPLOT, shown in Table 12. It uses CAL as
a subroutine for computing temperature and intensity parameter. The
results from INTPLOT are shown in Table 13. This same information
is plotted in Figure 21. The ordinate of the figure is the fraction
burned or as it is more generally termed, the combustion
efficiency. The abscissa is the intensity parameter.
Examination of the curves in Figure 21 reveal clearly
the nature of the critical conditions which normally exist within
a burner, and how, for some values of burner operating conditions,
these critical phenomena no longer occur. Referring to the curve
for equivalence ratio 1.0, we see that the ignition condition
corresponds approximately to INT = 371 which occurs at
about 10% combustion efficiency. This means that for an inlet
temperature of 2000°F, with a stoichiometric mixture, we will
achieve spontaneous ignition in the burner if we reduce the air
flow level to the point where the itensity parameter is below the
value 371 which corresponds to the ignition critical point.
The blow out condition, on the other hand, can be seen
from Figure 21 to occur at I = 982, at an equivalence ratio of
0.5. This means that once we have ignited the burner, we can then
increase the flow rate through the burner to approximately 2.5
times the flow rate at ignition before the burner will blow out.
Examination of the other curves in Figure 21 show that
these two critical conditions, which correspond to the vertical
tangents of the constant equivalence ratio curves, exist
for those curves for which the equivalence ratio is 0.3 or higher.
For 41 < 0.3, the curve has no vertical intercept. The curve for an
equivalence ratio of 0.1 does not even approach a vertical slope.
For this set of burner operating conditions, therefore, we cannot
speak of an ignition or a blow out condition. Some reaction
will take place within the burner at any value of flow
rate. In a manner of speaking, the burner is always ignited and
cannot be blown out. This is a particularly interesting phenomena
VII-19

-------
in view of the fact that the combination of high inlet temperatures
and low equivalence ratios which lead to this situation may
arise in burners for regenerative gas turbines.
It should be appreciated that the stability analysis
which was conducted using program BURN was not able to evaluate
stability limits when in fact none existed. To prevent the
computer from encountering this situation two safeguards were put
into program BURN. First, no efficiency values less than 0.3
were examined by the computer. Secondly , if the maximum value of
the intensity parmeter within the region of the analysis was
corresponded to the first point analyzed (E=0.35), the computer
skipped the remainder of the calculation and'went on to the next
point in the table of values that it was computing. The
conditions in Table 11 for which no critical phenomena occur
are indicated by *****.
VII-20

-------
SEMENOV'S THERMAL IGNITION THEORY
Figure VII-1
SIMPLIFIED COMBUSTION MODEL
Figure VI I-2

-------
VUL1S* COMBUSTION THEORY
T
Figure VII-3
ON CRITICAL PHENOMENON
Figure VII-4

-------
Q
LKFECT 01-' FLOW HATE ON CRITICAL PHKNOMENON	Kigure VII-5.
WITHOUT CRITICAL (IGNITION AND EXTINCTION) PHENOMENON Figure VII-6

-------
CHARACTERISTIC 'UPIIH'R T = V/° .05 D .85

-------

-------
V
p
CAN BURNER AND GUTTER BURNER CONCEPTS	Figure VII-9
VOL
(1 + K) W
C*- W
RECIRCULATION MODEL
Figure VII-10
BURNER ANALYSIS TECHNOLOGY

p tb



VOL

WA 6 W-
Ti

tb WB + W0 = wA + wF

6_ = FRACT. BURNED

INLET FLOW

EXHAUST FLOW
BURNER
Figure VII-n

-------
T
STK&UL OPEJVt\TmLOV4 OUT V\S_M" £>M_KNC£.	Figure VII-13

-------
THEORETICAL BURNER ANALYSIS
Figure VII - 14

-------
COMBUSTION INTENSITY PARAMETER - INT 			
Sec ft^atm^
Figure VII - 15

-------
THEORETICAL BURNER ANALYSIS
COMBUSTION INTENSITY PARAMETER - INT 	-	
Sec
Figure VII - 16

-------
THEORETICAL BURNER ANALYSIS
Figure VII - 17

-------
THEORETICAL BURNER ANALYSIS
a 
-------
THEORETICAL BURNER ANALYSIS
Figure VII - 19

-------
CUP! EP CTAHILITY Lit IT AII,'LYSIr
I:jTEI!SITv Pt RAIIt"TrP	ItiT = !'A/( VOL • P* " )
^	it T" 3ITY p/>r/»ri*~Ti r	h:t = ¦ A/(VL*r*~i
o
cori:usTinn n.'TrrsiTY p/'PAhrTrp Lor ;'l^t ti = 20on,
I f'TCf.S I TY PARAIlLTfr If.'T
1.00 3.1'' ia. 0" 31. c?. ioo. oc 31^ .23 loio.nr
IIJTfNSITY PARAMETER HIT

-------
VCALIU1
V	CAL
[1]	CA-0.27
[2]	CF+0.7
[3]	CB+-0. 32
[ 4 ] TBI+-TI
C 5 ~ HF*-( 0.02*2'J-537) + (0 . 0 004194x( (TJ*2)- 537* 2)) + (~7 . 921 8E~8x( ( ?!* 3 ) - 5 3 7 * 3 ) ) +6 . 4 32~ 1 2 x ( ri"*4 )- 537*4
[6] HB+HBO-( 0. 81+0.633) *HFxTHi 1 + 27/
[ 7 ] T2WP : TI+-TI
[8] TB+TI+ ( (HB*PH* (1 +TH)*E)- ( PP.xTH*HV'* 1-2 ) )*	( 1 + TH ) * E ) + ( CA x ( 1 -E*P1l) )+CFx p/'xJV/x i-£ )
[93 "~((! TB-TBI)Z1)/C0NT
[10] TBI+-TB
[11 ] CA+{ 0 . 2261 8 ) + ( 1 . HimE~SxTB + TI) + ~7 .82292~10x(ri?*2) + ( TExTI)+TI* 2
[12]	CP-«-(0. 02 ) + (0.0004194xTB+ri) + ( ~7 . 921 82"8X (rs*2) + ( 7Bx?I) + TI*2 ) + 6 . 43E~12x( (3-3*4 )- (TJ*4 ))i?B-TI
[13]	CB+(0. 28072 ) + ( 1. 52932 SxTP+TJ)+( 8.44582"10x(TB* 2) + (TBxTI) + (TI*2))
[14]	CB+CB+ ( 1 . 8508x( (TP*0 . 5 )-(.TI*0 . 5 ) )±TB-TI )-8B . 35B*TE-TI
[15]	+TEMP
[16]	CONT: TEI+-TB
[17]	INT+K* ( 1 -2)x27/x ( \ -E*PH ) x ( y,p i VA ) x ( *-EA iR*TP ) i ( 1 + 37.') * 2 x ( (pz/xr/Jxl-CJ + t ( 1 -E*PH) xpipiMA ) + P//xr* ( 1 + Trf ) TFifB ) * 2
[18 ] INT--INTx ( TB*0 .5)x(29x2116il545 xJZ?) * 2
V
[19]
•H
B-
H*
ft
<

-------

•7„'rr[r ]f

V
- . , ( pF
) )n ( ( pf ) pLY-^G ( ( P'1) p.
[11]
J ¦'-J +1

[121


[13]
"+!'+1

[14]
)// W

V
Table VII-2

-------
THEORETICAL BURNER ANALYSIS
Effect Of Equivalence Ratio and Efficiency On
Temperature and Intensity Parameter
Air Temp. = 0°F
TA
r. K,orv ¦
o. 0 00 0-"C
• ;. i... o
. . i'C.,„ (¦
0 . OC OC '"ij
~.orjcrc
¦j . cooo. ii
¦ .r:r:o rc
o. o o c> c; o
:. ooo j. n
o • c o o o
o.ooooro
o. nuooro
0.0000 0
o. ooooro
'j. ooooro
C.000070
0. 0000"', 0
0 . 00 00 "O
C . 00 0 0.7 J
C.0000/0
0.oocoro
o.ooooro
o.coooro
0.ooooro
o.ooooro
o. ooooro
o.ooooro
o.ooooro
c.ooooro
o.ooooro
o.ooooro
o.ooooro
o.ooooro
o.ooooro
o.ooooro
o.ooooro
o.ooooro
o.ooooro
o.ooooro
TA -
ETA -
INT -
PHI
1	. n c n 0 ~ 1
1.0000 "" 1
i.oooor"i
i. c o o d ¦' ~ i
2	. OuO'ir"!
o.oooo. i
. ooo or" l
".ooo;;r i
. :ooo; "i
3.00001"]
3. oooor"i
~.cooor"i
'I . 00 CO. ~1
4 . OOOO71"!
li. oooor"i
4.00 0 0 ~1
5.oooor"i
5.	oooor'i
s.rooor i
5.oooor"i
c. oocor"i
c. oooor"i
~.	oooor'i
6.	oooor"i
7.	oooor"i
7. oooor"i
7.oooor i
7.C000i"l
9.oooor~i
8.0000?~1
8.oooor~i
c.oooor~i
?.oooor~i
o.oooor i
o.oooor"i
9.oooor~i
l.ooooro
i.ooooro
i.ooooro
l.ooooro
ETA
'J . 0 :j u n ~ l
?. yonor"l
'J . 0 J (.  3
2 . 6 fi 7 3 l 3
2 . 01.' n "3
2.9301." 3
2 . 3 3 7 0 r 3
2	. 961ir3
3	. 2 3 0 9 r 3
3 . 2 57 71'3
3 . 2 & 0 M r; 3
3.2401 '3
3 . 5349r3
3 . 5643r3
3 . 5 6 7 2 r 3
3 . 505 5.", 3
3 . 8246r3
3 . R5G4Z73
3.8596"'3
INT
•>. "vior~n
"t n r f _ 1 1
-.: ? i :. ~ i:
"13
c.ri;r "f,
3. onm- r
. 2i5l ~7
4	. 0 7 " 3. ?
0	o 'j 7 *" ~ 2
•J. 5 4 09 ~4
i.oiicr 4
1	. ¦17 3 1^ ~ C
8.42 r,2;'~2
:.177G'-~2
2	. 3733""' 3
2	. 3 9 3 6 r — 4
h.99	44r~l
1.521ir"1
i.e	3C7r"2
l.C-4 1jr 3
2.0 7 34r0
5	. 3 02 ° '"l
5 , 0 8 6 7 r 2
5 . 717Gr~3
7 . 4 90 0r0
1 . 2 310. 0
1 . 2G00r"l
1 . 2649r~2
1.397iri
i.ssasro
1.9 24Sr"l
1.?283r 2
1.9334ri
1.9094ro
1.85G2r~i
1	. 8 5 03r~2
3	. a 10 3.-1
2	. -¦> 7 5 917 ~ 1
3	.112ir 3
3 . 12 5 9r~5
Air Inlet Temperature °F
Efficiency
Burner Intensity Parameter
PHI - Equivalence Ratio
TB - Combustion Temperature °F
Table VII - 3

-------
THEORETICAL BURNER ANALYSIS
Effect of Equivalence Ratio and Efficiency On
Temperature and Intensity Parameter
Air Temp, = 400°F
TA
4. CO00272
4. 0 0 0 052
4 . 0 0 0 01' 2
4 . 00 0052
4 . 000052
4 . 10005?
4 . 0 00 052
4.0000r2
4.00 0052
4. o o c or 2
4 .000052
4.000052
4.000052
4 . 00001'?
4.000052
4.000052
4 . 00T)0r2
4.000052
4.000052
4 .000052
4.00 0 052
4 . 00 00/72
4 . 000052
4.000052
4.000052
4 . 000052
4.000052
4. 000052
4 . 0000/72
4.000052
4.000052
4.000052
4.00 0 052
4 . 00 0052
4.000052
4.000052
4.000052
4.000052
4.000052
4.0000£2
PHI
i. oooo7~i
1 . 00nor"l
i. oooor"i
1	. oooor"i
2	.oooor"i
:.ooonr'i
2	. oooo.- "l
:.ooool"i
3	. 00005"l
3.	oooo5~i
3 . 00005~1
3	. 00005~1
4.oooor'i
4.cooo5~i
4	. 00 0 05"l
4. oooor'i
5	. 00005~1
5.00005 i
5.oooor"i
5	. 00 0 O'7'" 1
o. onoor~i
6	. 00005~1
6 . 0000~'~1
6	. 000 05"1
7	. 00 005" 1
7 . 00005"1
7 . 00 005~ 1
7.0 00 05 1
8.oooor"i
o.oooor"i
8.00005~1
8.00005"l
9.00005"l
9.000 05"1
9.00005 1
9.00005"l
1.000050
1.000050
1.000050
1.000050
ETA
9 . 0000,7"
'1. 9 00 0.7"
9.990AL~
9.99 2 07"
O.OOCOi"
9.y0097"
9.990 07"
9 . 9 9 i) 0?"
9 .0 00 07"
9 . 9CG07"
5.99005"
n 9 Q ° 0 7 "
°.C 0 0 07"
9.90005"
9.990C5"
9.9990, "
'). 0 00 0.7"
9.90 005"
°.9 9 0 05"
G.9 99 05"
9.00 C07"
9.90 005
9.9 90 05"
9.99905"
9.0 0 0 0 5"
9.90005"
9.99005"
9.9n9 05~
9.00005"
9.90005
9.9 90 05"
9.99 905"
9.00005"
9.9 0 0 05"
9.9900""
9.99 9 05"
9.00005"
9.9 00 05"
9.99005"
9.99 905"
TB
9.452452
R . PC0952
R. 9 34 0^2
C . 9 3 '.'0 72
1 . 2 7!! ~ 3
1 . 3 4 0 3 ' 3
1 . 34 Pf '3
1.340473
1	. l".4 0£53
1.7	5 j473
1.771^5'
1.77i453
1.9 99 353
2	. 13:, 753
". l r>; p 3
2 . ir. 7 353
2.335 :"3
2 . 517953
2 . 5 3 f> 1 ~3
2 . 5 3 7 9 J-3
2 . 6 5 3' 5 3
? . eG'¦ 3 53
2 . P B 5 3 73
2	. G07453
2.95 3453
3	. 19175 3
3.215 553
3 . 2 1 7 373
3 . 2 3 0 0 ""3
3.50 2 553
3 . 5 2 R £ r 3
3.5 315 73
3 . 508 673
3.79 8253
3.8271^3
3 . 830 053
3.76645 3
4.08	0 373
4.111753
4.11485°
INT
3	.005 'J5~C.
r .6 '. 89 5~7
7.164 'J5~9
7.219 o5"9
1 . 911 C'i5-3
" . i' 7fi 1 .~4
4	. 94'.35":,
4.8 7 f P~"G
7 . 2CA55"2
1. 53,35 """?
1.64'jr.5 3
1.GblO	5""
C.C9 2 55"1
1.34 215"l
1 . 42Uf>
1.433 V~ 3
3 . 1 6 7 r r0
5.50C41 1
5.79135"2
5.Pn0C5~3
8.9550"0
1.4	09 27 0
1 . 4 6 C 1 ~ 1
1.471?5"2
1	. 822251
2.5	00150
2	. 6 31 2 7" 1
2.C3825	2
2.8599-1
3.	44 34"0
3. 47377"!
3	. 4 7 6 4 5"2
3.491251
3.12745n
3.0147 5"1
3. 00257"2
2. 975551
4.4 9 99 5"1
4.67205"3
4.6094"~5
TA - Inlet Temperature °F	PHI - Equivalence Ratio
ETA - Efficiency	TB - Combustion Temperature °F
INT - Burner Intensity Parameter
Table VII - 4

-------
THEORETICAL BURNER ANALYSIS
Effect of Equivalence Ratio and Efficiency On
Temperature and Intensity Parameter
Air Temp. = 800°F
TA
PHI

ETA
TB
INT
P. 00 0 07 2
1 . 00007"
1
9.ooocr"i
1. ? 2 ^ 1; 3
1 . 413Sr"3
8.000072
1 .00007"
1
9.9ooor"i
1 . 2 6 3 9 r 3
2 . 12157~4
0.000072
1 .00007"
1
9.9°oor"i
1 . 26 8ir3
2 . 2 0 717"" 5
8 . 00 007.2
1 . 000027"
1
9.99 9C7~1
1.2GR:r3
2.215C7"?
8 . OOO.U'2
2 . 00007"
1
9 . OOOO.'~1
1 . 614 3r3
r.r 5 r ^ : "2
F . 00 00272
2 . OOOOff
1
9.90007 1
1.6939J3
1.1000 r"2
8 . 00 007?
2 . 0000^"
1
9 . 9900z "1
1. 7 01 P 27 3
1 . 1539 7"3
S . 0000272
2. oooor"
1
S . 9990r"l
1 .7020 73
1.15'. 57~4
8. 0000/72
3.oooor"
1
9. 0 0 0 07"1
1 . 9 8 0 77~3
7.2 310r"l
0. 00 00:72
3.oooor"
1
9 . 9ooor"i
2 .094 873
1.ir4:r"i
3.00 0022
3.oooor"
1
9 . 9 q c or 1
2 . 1 0 6 2 r 3
1.24047~2
8 . OOOOL'2
3.oooor"
1
9.9 9yor~i
2 .1 0 7 4 " 3
1.246ir"3
o.oooorr
4.oooor"
1
9. oooor~i
P. 3 24 373
3.5 34 3r0
G.OOOOT2
4.oooor"
1
9.°ooor"1
2.4 7 0673
5.54Z2r"l
8.0000:72
4.oooor"
1
9.99oor~i
?. 4 8 5 3
5 .777'ir :
fi. 00002:2
4.oooor"
1
9 . 9 9 90.""l
2 . 4 8 G G 7 3
:>. coi3r"3
8 . 0000272
5 .OOOOii"
1
9.oooor"i
2.6478.3
1.0 b 9 6 ;r 1
8.0 0 0 or2
5 .00007"
1
9 . 9 0 OOr, ~ 1
2 . 0 2 u 2 7 3
1 . 5C7ir.i
8 . 000072
5.oooor"
1
9.9 9 cor"1
2 . 841 9."3
1 . 62 35r"l
S. 000072
5 .00007"
1
S . 3 9 9 027" 1
2 . 8 4 3 C r 3
1 . 6292r"2
8 . 0000272
c .oooor"
1
n.0000',"l
T- . n5 3 5 ~3
2 . 2 0 5 5 r 1
8. 00 00222
o. oooor"
1
'J . 9 0 0 Of, " 1
3.15 0 4 r 3
3 . 1475r0
8 . 00 0 0272
6. oooor"
1
9 . 99 0 0.-7 1
3 . 1 7 0 f! 3
3. 2 3 6 3.1
G . 0000272
6.oooor"
1
9 .99 007"!
3 .1 r 0 r 2'3
3 ."4 5 1.7" 2
0 . 0000272
7.oooor"
1
9. oooor'i
3 . 2 '13 ir3
3.8782"1
8 . 3 0 00272
7.OOOO".-
1
9 . 90r0'1-l
3.475372
4.S4S7F0
8. 0 0 00.72
7.oooor"
1
'J . 9 S 0 0""" 1
3 . 4r;8 3. 3
4.93457"l
C.000052
7.oooor
1
9. 99 90; 1
3 . 5 0 0 r „ 3
4 . 9 4 2 9 7 2
C. 00 00772
c.oooor"
1
9 . 0000'" ~1
3 . 5 18:- 3
5 . 29 2 2 ~1
8. 0000:72
o.oooor"
1
9.9000 ""l
3 . 7 7 f j 3 '" 3
5 .79-970
O.OOOOZ72
8.oooor"
1
9.9900""1
3 . 8 02 073
5 . 7 0 P. 4 y, ~ 1
8 . 0 0 0 027 2
8.oooor"
1
9.9 990; "l
3 .804073
5 .790ir"2
8 . 0 00 0Z7 2
9.oooor"
1
9 .oooor'i
3 . 7 8 01 3
5 . 843271
8.0000T2
9.oooor"
1
9.9ooor 1
4. 063 3273
4.G237F0
8. 0 0 00272
9.oooor"
1
9 . 9900r"l
4 . 09 16.,'3
4.6167r"l
8 . 00 0 027 2
9.oooor"
1
9. 999 0""1
4.0944T3
4.5'jURJ 2
8. 00 0072
1.ooooro

9.0 ooor"i
4 . 0 3 C 0773
4 .611071
8 . 0000272
1 .0 00 0.70

9.9000r"l
4.337573
G . 4936.7"l
CM
O
O
O
C
1. ooooro

9.9 9 00r"l
u . 3 6 8 3 '3
6 . 7 00 37"3
8 . 00 00272
1.ooooro

9 . 9'J 90 7" 1
4 . 371 32:3
0 . 721ir"5
TA - Inlet Temperature °F	PHI - Equivalence Ratio
ETA - Efficiency	TB - Combustion Temperature °F
INT - Burner Intensity Parameter
Table VII - 5

-------
THEORETICAL BURNER ANALYSIS
Effect of Equivalence Ratio and Efficiency On
Temperature and Intensity Parameter
Air Temp. = 1200°F
TA
1.7000T3
1 . 2 C 0 CE 3
1.200073
1. 20007.3
1 . 2000173
1 .20 (10£"3
1 . 2 000E3
1.2000E3
1 . 2 00 0273
1.2000E3
1.2 00 0E3
1. 2000273
1.2000E3
1.2000E3
1.2000273
1.2000E3
1.2000E3
1.2000E3
1 . 20 00273
1.200023
1. 2000E3
1.2000273
1.2000E3
1.2000E3
1 .200 0E3
1.2000E3
1.2000E3
1.2000E3
1 .2000E3
1.2000E3
1.2000E3
1.200 OF 3
1, 2 00 Off 3
1 .2000173
1 .2000E3
1.200053
1.2000273
1.2000E3
1.2000E3
1.2000E3
PHI
1 . 0 00077" 1
i. oooor'i
1	. 0 00 0""!
1.0000"~1
2.oooor":
2.oooor"i
2	. 0 0 0 Of-1
2	. 000027~1
3	. 000 0£~1
3 . OOOOlTl
3 . 0 0 00E~1
3 . 00 0 0E~ 1
4.0000E~1
4.0000£~1
4.0000E 1
4. 000077~1
5	. 0 0 0 0E~1
5.0000E~1
5.0000E~1
5.00 00E~1
6.0000E~1
6.0000E~1
6	. 0 00 0E~1
6	. OOOOE'l
7.	ooooe~i
7	. 0000E~1
7.0000E~1
7.0000E~1
8	. 0 00 0E~1
8.0000E 1
8.	0 0 00E~1
8.0000E~1
9.0000E~1
9.	00 0 0E~1
9.0000£~1
9.0000E~1
1 .0000E0
1. OOOOEO
1.OOOOEO
1.OOOOEO
ETA
9.oooo:,_i
9.90C0"~1
r. . 9900E~1
9 . 999077 1
9 . 0000„"l
G . 3000~"l
9 . 910077" 1
9 . 9 9 9 0 Z7 — 1
'J . 00C077~1
9 . 90007;"l
9.99 OQE~1
9. 999 or"1
9. 000 OF-1
9.900 0 C~ 1
9.990 or 1
9. 99 9 0 27 — 1
9 .OOOOE'l
9.90 OOE-1
9. 99 0 0E~ 1
9.999 0"~1
9.oooor"i
9 .9000i7~l
9.990 or"1
9. 99 9 0Z7~1
9. 0 00 0E~1
9.9000" 1
9. 99 0 or"1
9.9990E~1
9.OOOOE~1
9.9000E 1
9. 990 027~1
9.9990E~1
9.oooor"i
9. 900 0E~ 1
9.9900E~1
9 .9 9 9 0E~1
9 . OOOOE~1
9 .900 0E~1
9.9900E 1
9. 99 9 0E~1
TB
1 .60 0 If3
1 . 6 3'J 
-------
THEORETICAL BURNER ANALYSIS
Effect of Equivalence Ratio and Efficiency On
Temperature and Intensity Parameter
Air Temp. = 1600°F
TA
PHI
ETA
TB
INT
i.fiooo :3
1.:ooov3
i . c o o o v 3
1 .600 0/; 3
1 . f 000""3
1 .6 00 023
1. C CO 0273
1 . C00023
1 . 6 00 01:3
1 . G 0 00 ^ 3
i. r,ooo73
1.000073
1 . 60 0023
i .ecoo.'3
1. 60002 3
1. G 0 0 Off 3
i . 600073
1 . 6000173
1.GO 00 73
1 .0 00 0/73
1 .600023
1 . 600023
1. 600023
1 .600023
1. 600023
1 .6000/73
1 .600023
1.600023
1.600023
1 .600023
1.600023
1 . 60 00273
1.600023
1.600023
1.600023
1.60 0023
1.600023
1.600023
1.GO 0 023
1.600023
00 00 2
00002 '
0000"'
oooor"
2. 0000.7
2	. 0000"'
2.00007"
r.oo o or"
3	. 00 0 0""
3.oooor"
3 . 0000"'
3	. 00 0 0.7'
4.ooocr"
4	. 00002"
4	. 00007"
4.00002"
5.00002"
:.oooor"
5	.oooo;:'
5.0 00 02'
G.00002'
r>. oooor"
e.oooor"
C.000 02"
7.00 0 02"
7.oooor"
7.000 02"
7.0 00 02"
8.000 02'
8.00002"
0.00002'
o.oooor"
9.00002'
9.00002"
9.0 00 02"
9.00002"
1.000020
1.000020
1.000020
1.000020
0 mi p *
9 0 C 0 _
oooor
gq ?c:
9.0000'
9.90C02"
M.9°002"
c. :'3
2	. 0 21473
2	. 3 3'". 7?
2.1"7"23
2.414573
2.	i-l 52. 3
2.603473
2.7735 3
~ .781073
2.7R502?
:. 9G3'.:"1
3	. 11.97 "3
3 . 1 32273
3.13 3 e _3
3.2	TO,23
3 . 4 4 5 2 7" 3
3 . 4 016 73
3 . 4 6 3 2 713
3.	5 03423
3.755123
3 . 7 7 4 2 27 3
3.77 6123
3	. 831'"•73
4.043923
4.07177?
4.073973
4.007723
4.331123
4.3	55 523
4.3 57973
4.331623
4.5908.3
4.626773
4.62°423
4	. 5 648273
4.857373
4.G8C	77 3
4.889 723
10 CI7' 1
r i : < 2 " l
0 2 717""
0 2 03.7" 3
u . P200
5 . 8 16 7 2 ~ 1
5.024 l.~~2
5 . C 3 4 n .7 ~ n
1.5 36 271
1 . 8-J 7 010
1 . 935r.2"l
1 . ?3-4""2
3 . 5 17 "7 1
4.3171"0
u . 400"2~1
4.40272"2
6.377521
7.632370
7.756G2"l
7 . 76n02""2
9 .684721
1 .110021
1.122520
1. 1 23 72"" 1
1.2669^2
1.3 55 921
1.36 0 220
1 . 360 62~1
1.435072
1.36^121
1.347620
1.345 92"1
1.36 69 22
9.952020
9.41952~1
9.364 5 7~2
9.603121
1.208820
1.23502"2
1.23767~4
TA - Air Inlet Temperature °F	PHI
ETA - Efficiency	TB
INT - Burner Intensity Parameter
Equivalence Ratio
Combustion Temperature °F
Table VII - 7

-------
THEORETICAL BURNER ANALYSIS
Effect of Equivalence Ratio and Efficiency On
Temperature and Intensity Parameter
Air Temp, = 2000°F
TA
2.000023
2. C 0 0 OA13
2 . 00002:3
2. 000 023
2.000023
2. 00 00/73
2.000023
2.000053
2.000023
2 .000023
2.000023
2. 00 0C23
2.000023
2.000023
2.000023
2.000023
2.00 0023
2.000023
2.000023
2.00002 3
2.000023
2.000023
2. 000023
2.000023
2.000023
2.000023
2.000023
2.000023
2.000023
2. 000023
2.000023
2.000023
2.000023
2.00 00 23
2.000023
2.000023
2.000023
2.000023
2.000023
2.000023
TA
ETA
INT
PHI
1 . 0000F"l
1.00 002~1
1.00002"1
1.00002"l
2.00002"1
2.00002"l
2. oooo;;~i
2.00002 1
3 . 00 002"1
3 . 00 002~1
3 . 00002"l
3	.0 000i'"l
4	. 00002~1
4.0000E"l
4.00002"l
4.00 002~1
5.000 02"1
5	. OOCOi: 1
5.00002~1
5.0000f~l
6.oooor"i
6.0 0 002"1
6.00002"l
6. 000 02"1
7.00002"1
7.00002"1
7 .000 02~ 1
7.00002"l
8.0000E~1
8.00002 1
8.00002~1
8.00 002"1
9.oooor'i
9.00002"l
9.000 02"1
9.00002"1
1.000020
1.000020
1.000020
1.000020
ETA
? . 00002"l
9 . 90002~1
9 . 99002~1
9.99902"1
9.oooor~i
9.9000r"l
9.99002"1
9. 999 P2~ 1
9.00002~1
9.9ooor~i
9.99002"l
9.99C02~1
9. oooo2~i
9 . 90 0Oi. 1
9.99 002"1
9 . 99902"l
9.00002"l
9.90C02~1
9.99002~1
9.999C'2~1
9 . 0 f) 0 0 2~ 1
9.90002"1
9.99002"l
9.999 02"1
9 . 0 0 0 0 2~1
9 . 90 002"1
9. 9 C 00 2~1
9 . 99902"l
9.00002"l
9.9 o o o;" 1
9 . 9900r"l
9.9 9 9 02"l
9.00 0 02"1
9.90C02"l
9.99002"l
9.99 902"1
9.0 00 02"l
9.90 002"l
9.99002"!
9.99902"!
TB
2 . 359923
2 . 395923
399523
2.399023
2 . C- 9 7 323
2. 7672 '3
2.774173
2	. 7748." 3
3	.016 323
3 .116:'23
3.127023
3 . 12 '10 ¦ 3
3.317423
3.4475-1
3 . 4 6 1'4 l 3
3.46172T
3 .602723
3.7008. 3
3.77r 6 "3
3 .778223
3	. 8 7 3 8 " '3
4.058723
4.077223
4.0791' 3
4.131823
4	. 342f-.73
4 . 3 C 3 723
4.36 5 823
4.377 723
4.613723
4.637423
4.639823
4.CI 2423
4 . 8727 ^3
4 . 8 9 892 3
4.9 01523
4 . 8 36 923
5.124123
5.1£3123
5.156023
INT
6.016620
6.29512 1
e.3?512"2
6.32 822"3
1.922321
2 . 141 r)20
2.1645L~1
2.1668" 2
4	. 4 ? 2 6 21
5	. 0 49 32p
5 .11292"1
5.11912 2
8.13.721
9 .25': 520
'J . 3 7 1 52~1
9 . 3C2 72"2
1.201622
1 .U0FI021
1 . 421720
1 .4 2312"1
1 .705922
1 .C 3 4 721
1.84 4P20
1.84 5 5 2~1
2.043122
2.063821
2.059520
2.0590 2~1
2.1F0422
1.94 8621
1.915920
1.91252"l
1. 94 CI22
1.3517^1
1.2 7 3820
1.2C592"l
1.309 322
1.581320
1.609 72"2
1.C1252"4
-	Air Inlet Temperature °F PHI - Equivalence Ratio
-	Efficiency	TB - Combustion Temperature °F
-	Burner Intensity Parameter
Table VII - 8

-------
VSTABILITYtD]V
V	STABILITY\I
[1]	TIl*H60+ 0 400 000 1200 1600 2000
[2]	ANSR*OpO
[3]	1*1
[4]	TI*TILiI1
[5]	BURN
[b] ANSR*ANSRA.ANS)
[7]	' '
[8]	1-1+1
[9]	-f(J"£priL)/4
V
Vtfi/fltf[L|]V
V	BURN;J jJ
[1]	12 5 pO
[2]	-~(PRItlT = 0)/SkIP
[3]	'	TI = ' ; (2V-460) ; ' of"
[4]	1	Ptfi"	ETA	TB	HIT	1
[5]	SKIP: PH >1-0. 1* i 10
[6]	J*1
[7]	JUP;PH*PHAtJ]
[8]	COUNT*- 0
[9]	E*0.3 + 0.05 x i 1M
[10]	ITT-.C0Ul!T*C0\JNT*\
[11]	CAL
[12]	INTM*[/INT
[13]	I-INT\I HTM
[14]	-+(1 = 1 ) /OUT
[15]	¦*( COUNT = 3) / OUT
[16]	£-E'[i-i] + (o.i*coyjvr+i)x»io
[17]	-jzt
[18]	OUT: +(PRINT-Q) / STOR
[19]	12 3 12 4 12 1 12 6 DFT PH,£[I],(TBI I]-460),INT[I]
[20]	STOR :ANSt J ; ]«-( (TI-4 60 ) ,PH ,E[I] ,(TBlI]-HS 0 ) ,imi] )*I* 1
[21]	cW + 1
[22]	+(JZ10)/JUP
V
Table VII-9

-------
THEORETICAL BURNER ANALYSIS
Stability Limits
TI = 0
phi
0.100	0.9480
0.200	0.9390
0.300	0.9250
0.400	0.9100
0.5 00	0.8:60
0.6 00	0.87 9 0
0.700	0.8500
O.GOO	0.0?'JO
0.900	0. 3(.00
1.000	0.7600
OF
rt	t
J .	J I' -
4 9 4.1	0.000000
940 . 2	O.OOOC'IO
133G.3	0.003139
1G95.4	0.08UQ05
2019.fi	0.700100
2 3 0 0.2	2 . 936 163
95 35 . 6	fi.03^965
2	7 4 :. 7	1C. 408577
? c 0 5 . R	?7.0P3e6 5
3	0 0 4 .9	3 0.1Z6 * 0 ,
TI = 400
Pill	ETA
0.100	0.8300
0.200	0.8980
0.300	0.8900
0.400	0.8770
0.500	0.86C0
0.600	0.8450
0.700	0.8200
0.800	0.7990
0 . 900	0.7f 60
1.000	0.7280
OF
T7	II.'T
04 0 . 4 0 .000003
1255.5 0.001949
16 2 7 . C 0.07304C-
1960.5	0.715631
2254.6	3.3 4 733"
2523.7	9.05005"
2740.1	21.800524
2939.1	38.516837
3074.3	57.941062
3162.3	77.320775

TI = 800
OF

Phi
r.TA
j1 L
ii:t
0.100
0.7900
1170.8
0. 001762
0. 200
0.8380
1559.3
0.074312
0.300
0.8400
1904 .2
0.GO 5 49 8
0.400
0.8300
2209 . 8
4.038581
0.500
0.8200
2490.0
12.647128
0.600
0.8000
2724.2
29.034448
0.700
0.7000
2931.6
53.657546
0. 800
0.7550
3099 . 8
84 . 55 8451
0.900
0.7200
3210.6
118.108823
1 . 000
0.6870
3299.2
150.905892
TI -
• Air Inlat Temperature °F
PHI ¦
- Equivalence Ratio
ETA ¦
- Burner Efficiency
TB
¦ Combustion Temperature °F
INT -
Combustion Intensity Parameter
***** .
- No Critical Conditions Arise
Table VII - 10

-------
THEORETICAL BURNER ANALYSIS
Stability Limits
; /it	i:ta
o.ioc	0.5'<00
0.200	0.7480
0.300	0.7700
0.400	0.7700
0.500	n.7-00
0.60C	0.7400
0.700	0.7280
0. 8 00	r, .70 10
0.900	0.67"0
1.000	0.63 0 0
1°0 0 or
rr	ii t
14C2.3	0.126897
inn4. 3	1.11817 0
21ru . 0	5 . 476 255
244 7.3	17.233346
2695.8	40.177244
2919.4	75.838458
3102.9	12 2.01519 7
3 i>. a	176.558030
3347 . 2	232. 570C4 9
3 4 2 r> . 2	2 86 . 2 3913'
TZ =
>'hi	ETA
0.100	iiMthit
0.200	0.5000
0'. 3 0 0	0 . C 5 7 0
0.400	0.CQ70
0.500	0 .G 0 6 C
0.600	0.5760
0.700	0.C590
0.800	0.5360
0. 9 00	r>. c010
1.00 0	0.5 800
ICOO OF
rp "	t- 7i' r\
sVfti'.ftft	hitititit
2074.2	9 . 7 2 4 C18
2394.4	27.133C38
266C.3	60.721746
2887.9	113.124843
30 8 3.6	183.1832 56
3245.6	266.406383
3 3 7 2 . 1-1	356 . 503031
34 0 5.?	447.141329
3530.4	533.266681
?I =
"VT	ETA
0.100	ititititit
0 . 200	it it it it it
0.300	0.4060
0.400	0.54C0
0.500	0.5700
0.600	0.5700
0.700	0.5650
0. 800	0 . 54*50
0.900	0.5270
1.000	0.5000
2<">0 0 OF
r7	Il'T
ftftiVjl:':	A
ititititit	ititititit
2458.5	113.805829
2801.9	190.50S831
3019.4	295.5 9 7717
3192.6	422.142911
3 3^5.7	562.614364
34 5 0.8	708.456281
3540.1	851.956971
3587.8	988.108951
TI
- Air Inlet Temperature °F
PHI -
- Equivalence Ratio
ETA -
¦ Burner Efficiency
TB
- Combustion Temperature °F
INT ¦
• Combustion Intensity Parameter

¦ No Critical Conditions Arise
Table VII

-------
VIJITPLOTZ G]V
7 HITPLOT
[1]	HO+1
[2]	PINT*OpO
[ 3 ] PE*-opo
[4]	PUN*0.IX 110
[5]	E- 0.01 0.025 ,0.05X120
[6]	ITTiPH-PHNLKOl
[7]	CAL
[8]	PINT-PINT,I!1T
[9]	PE-PE,E
[10]	NO-NO+1
[11]	-KA'OSIO )/ITT
V
Table VII

-------
THEORETICAL BURNER ANALYSIS
cofx,jriTi'"¦i1 I'TfitiriTY p/*rArnrr iitcti, r, pi-o
TI - 2000 F

PH> {
0.1
0.2
0.3
0.1
0.5
0.6
0.7
!
1
! 0
•
CO
0.9
2.0
0.011
[
921.08
022.1'
°23.7
9 2'l. 79
;23.7'.
n.5r
"27.2'
-"27.83
0 2 P.2n
028 6r
0.025
1
112.02
126.53
i:;.7:
ll r ^ r -?
"61.89
17'.f"
188.""
Jlno -
51".2°
:ro;rr'
0.05
r
i
212.15
?30."3
r'15.1-;
2r7.cil
2-6.7r
305.66
321.
3.r i
3r2. 3''
^80.8 3
0.10
V
i
112.07
131
irc.o:
18-". "8
211.88
211.23
272.18
3" . r 0
337.8
371.3-2
0.15
73.'123
102.7:
131.25
16 3.92
2^0.6'i
P '! 1. P 7
^85.6 S
3 3 3 i 0 ¦'
383.31
13".13
0.20
\
61.101
87.6''6
120.13
1 r " . 0 2
20'.19
23?.11
318).' 6
383. 1'J
'152.87
526.-1
0.25
i
50.978
7 C.n. 21
i]n.r-
16 2. ">2
218.61
281.71
3r0.11
3-33
333.66
62°.73
0.30
[
'13.809
73.227
111.13
lr7.5.
231.15
313.69
''05.12
307.17
627.51
731.15
0.35
i
38.151
r?.l
113.73
173.95
0 r; 0 r t
t ^ ~ ^ 1
3M.17
1 ^0, J|n
569.31
607.il
830.28
0.1)0
•
j
31.192
65.783
113.7r
180.25
266.15
370.76
l^l!83
621.7r
7r5.53
9C8.o3
0 . '15
t
30.616
62.8,2"
113.r
l8r.C
2 79.6
3:1.27
5?r. 15
668.82
8I6.I5
961.71
0.50
*
,77.18
50.021
112.33
181.21
289.5°
"11.61
550.16
r3 7.
8'11. 50
032.16
0.55
i
L
21.621
5*.8rr
121.nf'
in r, .31
291.8;
'121.01
r-f 2.12
7n7.1'1
81^.05
967.07
o. ro
I
21.930
r3.17")
207.°1
188.''7
29I.12
12°. 75
35°.^?
6°5.71
pl8.72
921.33
0.65
n
I
i
1°.313
1n. 62
lr3.11
182.85
287.17
10 0 .r 2
53°.72
661.^8
762.1"
826.95
0.70
16.78
k-.irr
"6. 0'':6
172.;i
272.2^
386.35
r03.' 6
rnr.ii-'
/-7n.H
707.22
0.75

11.201
39.99?
87.337
158.25
219.1
350.69
I50.il
52f.8"
r. 7 3. 6 3
565."
0.30
i
I
11.37C
3'i.0ir
75.810
138.32
217.13
302.16
3*1.51
r 2
¦'1r 117 ?
''13.13
0.85
\
0.8667
27.1?
ci ,-ni'
112. 7*1
17r.12
?i! 2. np
nr H y 0
¦5 ? n -7 <'
/¦> <5
; - c • ' r
2 ' 2.71
0.90
*
r
( .nhq2
1".212
2^3
Pl\\22
2 2^. or
17- j, P
ooi. £
21r .'' 7
nor
2 30.^3
0.95
i
L
3.1005
1n.201
23.R:
Jl 0 7,'"?
fi J • f 1
r7.1r2
89.122
3 0 3*^2
10'.rn
" 3! 0 2 3
3 ^ . l| 1:r"
1.00
*
n
n

n
0
A
p
0
n

TI	-	AIR INLET TEMPERATURE °F
INT	-	COMBUSTION INTENSITY PARAMETER
PH	-	EQUIVALENCE RATIO
E	-	COMBUSTION EFFICIENCY
Table VII
- 13

-------
VIII. CORRELATION OF THE EXPERIMENTAL DATA WITH A THEORETICAL MODEL
The experimental data obtained in this program would be
of value by itself without additional theoretical analysis, since it
shows the remarkably low emissions which can be obtained by
proper operation of the Paxve burner. The value of this data,
however is immeasureably increased by our ability to interpret
it in the light of the theoretical burner analysis presented in
Secion VII of this report. We have been able to use this
analytical background as the basis for data correlations. With
those correlations, we can predict performance of burners having
widely different sizes and operating conditions. Such correlations
and the theoretical understanding of burner operation are of
particular value when new operating conditions are of interest.
The Paxve burner tested here had a maximum air flow rate
of 180 lbs/hr. The highest flow rate normally tested was
approximately 140 lbs/hr. At the nominal operating condition for
the burner (f/a = 0.038) this yields aporoximately 100,000 BTU/hr
of heat release. The volume of the burner was on the order of
0.03 ft3 and hence the nominal heat release rate was approximately
3 x 106 BTU/hr. ft3.
The main thrust of the EPA program, of which this
research work was a part, deals with burners of considerably
larger heat release; on the order of 2,50 0,000 BTU/hr. The
first task of data correlation, therefore is to be able to
predict the influence of increasing the scale of the burner by
at least a factor of 25. There is also a great deal of
interest at the present time in low emission burners for
application to gas turbine engines. Gas turbine burners
operate at higher inlet pressures and temperatures than the burner
under test here. It is therefore desirable to have a means for
designing a burner which will have low emissions at elevated
pressures and temperatures and to be able to predict the
influence of these parameters as well as scale size on both burner
stability and emission characteristics.
In order to accomplish the goals outlined above, several
types of experimental data correlations were investigated. The
stability data v/as correlated with the theoretical prediction
of the burner theory outlined in Section VII. The correlation
is presented herein. The carbon monoxide and hydrocarbon emissions
from the burner were correlated in terms of the efficiency
predictions of the burner theory outlined in Section VII above.
Correlation of the oxides of nitrogen data is discussed herein. The
combustion temperatures used in that correlation are those predicted
by the combustion theory.
A. Correlation of the Experimental Stability Data
Figure 20 of Section VI, shows the experimental blowout
data from the Paxve burner plotted against air flow; this data can
be compared directly with the predictions of the burner theory

-------
discussed in Section VII. In order to accomplish such a
comparison, a computer program, PREDICT, was written which
examines all of the lean burning experimental runs in terms of the
burner theory outlined previously.
1. Program PREDICT
Program PREDICT is shown in Table 1. An
examination of this program will be helpful in discussing the
experimental data correlations which are to follow. Because of
the limitations of the analysis, only equivalence ratios of 1
or less were considered. For each of the lean runs, a value
of the combustion intensity parameter, INTD, at which the burner
was tested was computed. Using the burner air inlet temperature
and equivalance ratio, a subroutine, Program LIMIT was called.
Program LIMIT, shown in Table 2, finds the limiting
value of the intensity parameter INTL corresponding to the stability
limit of the burner. Program LIMIT in turn called CAL, discussed
previously, which finds values of combustion temperature and
combustion intensity parameter as a function of the burner
efficiency. Program LIMIT performs essentially the same
calculation as that performed by Program BURN described previously.
Program LIMIT returns values of the limiting intensity parameter,
INTL and the efficiency and burner temperature which exist at the
blowout limit, EL and TBL.
Program PREDICT now 'compares the intensity parameter
at which the burner was operating with the value of the intensity
parameter at the stability limit. A stability prediction
parameter is formed given by
INTR = INTD
INTL
INTR greater than 1.0 implies that the burner will not stay lit.
The air flow rate and hence INTD exceeds the limiting value for stable
operation. INTR less than or equal to 1.0 implies that stable
burner operation should occur.
We must now compute the operating efficiency and burner
temperature. To accomplish this Program PREDICT uses, EL, INTL,
and INTD to make an estimate of the operating efficiency of the
burner. It then calls Program CAL and uses this estimated
efficiency to compute new values for burner intensity parameter
and operating temperature. The new value of intensity parameter,
INT, is compared with IHTD and the estimate of efficiency
revised. The process is iterated until the value of INT agrees
to within 1% with INTD. The efficiency EFF and operating
temperature TP of the burner are now stored for this run and the
next set of operating data is examined. The process is repeated
until all of the data has been exhausted.
The results of Program PREDICT are seen to be of several
types. First, we obtain a value of INTR for every lean run, and
VIII-2

-------
hence a prediction of whether or not the burner will be stable.
Secondly, for those conditions where the burner will be stable/
we obtain values of EFF and TP: the predicted efficiency and
operating temperature of the burner. Finally, for all runs
including both stable and unstable runs, we obtain values of
the efficiency and combustion temperature at the stability limit.
The use of Program PREDICT for comparison of burner
stability limits with theory requires some ingenuity as to how
one should present the data. In this regard it was decided that
a plot of the experimental observations of burner operating
condition versus INTR would be of some interest. A variable
named BURN was devised to assist in making this comparison.
For those runs in which the burner was considered to be operating
normally, a value of 2 was assigned to BURN. For those runs for
which the burner was definitely going out in a reasonable period
of time (less than five minutes), a value of 0 was assigned
to BURN. Those runs for which the burner was operating in an
erratic fashion, or for which the burner eventually went out
after a long period of time, were assigned a value of BURN equal
to 1.
Figure 1 shows a plot of BURN versus INTR. The upper
line represents conditions for which the burner was stable. The
middle line represents stability limit operation. The bottom
line represents test runs for which the burner was going out.
Clearly the assignment of a value to BURN is a matter of [judgment,
and hence it is not surprising that there is some scatter in the
data.
Figure 1 shows a surprisingly good correlation for
the predictions from the values of INTR and the BURN = 0 and
BURN = 2 lines. Virtually all of the BURN = 2 values lie below
INTR = 1.0. Of the runs for which BURN = 2, only 12 of these
have values on INTR greater than 1. The highest is run No. 223 for
which INTR = 1.96. Similarly, of the runs for which BURN = 0
was assigned, only one run, No. 391 had a value of INTR below
1.0. For run No. 391, INTR = 0.84.
The runs for which BURN = 1 was assigned do not show
quite such good correlation. Here there are lean limit runs for
which INTR = 0.25 and other lean limit runs for which INTR = 2.5.
Most of these lean limit discrepancies appear to be a result
of the difficulty associated with obtaining accurate fuel/air
ratio data. While the volumetric gas analysis equipment is
quite accurate for normal burning runs, it becomes less
valuable near a blowout condition. When the burner is near
blowout, the efficiency of the burner may drop to as low as
90%. Under these conditions, the fuel/air ratios based on the
volumetric data should be in error by about 10%. While this
may seem like a small error, a change of burner temperature of
200°F has a pronounced effect on the computed value of INTR.
2. Stability Tests From Program PREDICT
The stability predictions obtained from
VIII-3

-------
Program PREDICT are tabulated in Tables 3 through 18. The first
column in each of these tables shows the run number. Next we see
volume of the burner under test, then the inlet temperature and the
air flow rate. In Column 5, the equivalence ratio at which the
burner was operated is shown. This is based on the nominal fuel/air
ratio discussed previously. Column 6 shows the burner intensity
parameter based on the flow rate and volume of the burner.
Column 7 shows the limiting value of intensity parameter at lean
blowout as computed from the equivalence ratio and air inlet
temperature. From these the stability parameter INTR in Column 8
is computed. Column 9, labeled BURN, shows the estimate of burner
stability provided by the burner operator or the data in the
burner notebook. As explained above, 2 represents stable operation
of the burner, 1 represents lean limit operation and 0 corresponds
to operating conditions for which the burner will go out.
Examination of Table 3 through 18 provides additional
insight into the validity of the burner stability theory. We see
that in many instances where it was difficult to judge how close
we were to the stability limit, the value of INTR is in fact close
to 1.
Another interesting observation from the Table deals
with those values for which the burner was at or near lean limit
ODeration while the value of INTR is low. Here, the computer
program predicts that the burner will have some considerable margin
of stable operation, while m fact it was near or at its blowout
limit. These were usually runs at low air flow rates in test
Stand 2. This burner is more subject to heat loss than the other
burners tested. The influence of heat loss on the stability of the
burner can be significant, particularly at low air flow rates.
Since the burner stability analysis program did not take this
heat loss into account, it is not surprising that there are some
differences between the prediction and the experiment at the low
flow rates.
3. Correlation of Stability Data with
Theoretical Stability Curvis
In addition to the information obtained from
Program PREDICT, the stability limit data previously obtained from
Programs STABILITY and BURN can be used to correlate burner blowout
data.
Figures 20, 21, 22, 24 and 25 in Section VI
show blowout data for the Paxve Burner. Superimposed on these
figures are theoretical stability limit curves based on the analyses
of Section VII. It is clear that the theoretical limit curves
agree well with the experimental data. The propane curves are more
complete than the kerosene curves because it was easier to run the
propane tests. Figure VI 20 is for propane at ambient temperature
in a 33 cu. in. burner. This figure shows that all of the blowout
points (squares) lie to the left of the theoretical limit line,
while all but two of the stable burning points (circles) lie
to the right of the line. The lean limit points showing marginal
stability (triangles) are scattered around the theoretical limit
VIII-4

-------
line.
Another way of showing the same information is to plot
the data on a curve of equivalence ratio versus burner intensity
parameter. Figure 2 shows such a plot for the ambient temperature
data. Superimposed on the data is the theoretical limit line for
75°F inlet temperature; we again see that the data and the
theory are in substantial agreement.
4. Final Comments on Burner Stability
Correlation
The experimental stability data from the Paxve
burner agrees quite well with the predictions obtained from the
burner theory, particularly at high flow rates. There are several
matters which bear further investigation.
a.	The stability of the burner is particularly
sensitive to extraneous heat loss. Incorporation of a heat loss
term in the stability analysis should improve the correlation
between the experimental and theoretical predictions, particularly
at low flow rates. The interpretation of such a burner heat loss
parameter in terms of burner construction considerations would be of
great value in the improvement of burner design.
b.	The burner stability prediction program
assumed that the fuel entering the burner was at ambient
temperature. A correction was made for the heat input necessary to
raise the fuel to the air inlet temperature. During many of the
runs, particularly those with kerosene, the fuel was at an
elevated temperature. This seems to increase the burner stability.
The burner stability prediction program could be readily modified
to take into account the fuel inlet temperature.
c.	The burner analysis was limited to lean
operation. This was primarily a result of the limited funding
available for this effort and a corresponding limited interest
in this area of burner operation for automotive application. Some
industrial processes utilize staged combustion with rich mixtures in
the first stage. An extension of the present work to investigate
rich operation may reveal procedures that would assist in reducing
the level of pollutants emitted from these sources.
d.	The stability of the burner as well as its
emission characteristics are influenced by the uniformity of the
fuel/air distribution within the burner. Burners which have highly
homogenous fuel/air mixtures at the inlet are generally somewhat
less stable than burners which have non-uniform fuel/air distribu-
tions. This is due to the fact that the flame can stabilize in a
locally rich portion of the flow and then spread through the rest
of the stream.
Non-uniform fuel/air distribution is also a factor in
burner emissions. The problems of hydrocarbon emissions from the
top of the vapor generator stack, which caused so much trouble during
VIII-5

-------
the course of this program, was finally traced to a badly distorted
fuel/air profile in the air/fuel mixture ahead of the burner inlet.
Once this non-uniform fuel/air condition had been corrected, the
hydrocarbon emission problem was immediately cleared up.
Unfortunately it is not possible at this tine to go back and
establish a uniformity of the fuel/air mixture during the early
stability tests.
If the indicated improvements in the burner theory were
made, it seems clear that the theoretical analysis will be more than
adequate to serve as a basic tool in burner design and development.
The burner stability theory predicts the influence not only of
flow rate and burner volume, but also air inlet temperature and
pressure. Verification of the validity of this analysis at high
inlet pressure and temperature is a task of great importance. That
work was beyond the scope of this program.
B. Correlation of the Oxides of Nitrogen Data
It is clear from an examination of the experimental
data that the oxides of nitrogen emissions from the Paxve burner ex-
haust are a strong function of fuel/air ratio and not strongly dependent
on air flow rate through the burner. An examination of comparable
data from cold air and hot air runs shows somewhat higher NOx levels
at the elevated inlet temperatures. Although the data scatter makes
an exact comparison difficult, it appears that the influence of
inlet temperature is what one would expect if the NOx were a function
of combustion temperature only. Figure VI-19 shows all of the NOx
data from the burner plotted against burner temperature. Despite
the scatter it is clear that a correlation between these two
variables exist.
It seems reasonable to presume that the oxides of nitrogen
in the burner exhaust are formed in the burner by a chemical
reaction whose rate is given by an Arrhenius tyoe equation
d_(NO)_k[Oo]a Vol e~E/RT
dt	^
where: NO = Wt of MO fornec!
Vol = burner volume
E = activation energy of the over-all reaction
Since the NOx concentration in ppm is the ratio of the NOx weight
flow to the air weight flow, we might expect that the NOx in
parts per million would be given by an equation of the form
[NO] = K[02]a e-E/PT
WA/Vol
Examination of the data, however, shov/s that the predicted
inverse dependence of NOx with air flow rate either does not exist,
VIll-6

-------
or else is suppressed by the data scatter. The predicted
dependence on temperature however, is clearly evident, and can
be further characterized by plotting the NOx concentration against
reciprocal combustion temperature.
Figure 3 shows such a plot for all of the data. Figure
4 shows a similar plot for the small sample of the data in which the
fuel type and general operating conditions were relatively fixed. In
both cases, we can draw a straight line through the data and obtain
an empirical correlation.
Paxve has examined several approaches to the
correlation of the NOx data obtained during this program. The
possibility in involving the air flow, and the oxygen concentration
in the result was investigated. We found that a better correlation
could be obtained by using the combustion temperature alone than
could be obtained by including these other factors. The ambient
temperature data appears to be well fitted by an equation of the
form
NOx = 4.38 x 10s x e-E/RT
with
E = 36.7 K cal/mole
Figure 5 shows this equation superimposed on the summary NOx
curve, using the theoretical flame temperatures for 70 °F and
400°F inlet temperatures. It is clear that the curves give
a reasonable fit to the data, although the 400°F curve is
somewhat conservative as compared to the bulk of the high
temperature data.
C. Correlation of the CO Emissions Data
The theoretical burner analysis described in Section
VII of this report provides a basis for correlating the CO
emissions from the burner. The theory yields values for the burner
efficiency as a function of the inlet temperature and the burner
intensity parameter. If we substract the efficiency, from
1.0, we obtain the unreactedness, <5, which should be proportional
to the emissions of unburned or partially burned material in the
burner exhaust. Program PREDICT gave an evaluation of each lean
test point. Tables showing the predicted burner efficiency,
the predicted emissions, and the acutal values of the CO, HC, and
NOx emissions, are presented in Tables 19 through 31. Examination
of the data in those tables show that the CO emissions are generally
less than the predicted values.
In order to examine this further we have taken some
typical sets of CO gm/Kg emissions data and superimposed the
predicted emission levels based on the analysis. Figure 6 shows
the CO emissions from the burner for the small burner (33 CU. IN.)
on test stand 1 with ambient air and propane. The CO emissions are
obviously strongly influenced by the air flow rate. Curves
VIII-7

-------
through the data for air flow rates of 25 lb/hr, 50 lb/hr, and
100 lb/hr have been drawn on the figure as solid lines. Predictions
based on the theoretical burner analysis are shown on the figure
as dotted lines. Two of these theoretical prediction lines are
shown, one for 25 lb/hr, the other for 100 lb/hr.
Figure 7 shows a similar set of data for later runs in
test stand 2 burning kerosene with elevated inlet temperatures. Once
again, the predicted values are substantially higher than the
experimentally determined CO emission levels.
We do not have a simple explanation for the low levels
of CO emissions observed. We suspect, however, that this is a
consequence of the oversimplification involved in arriving at
the predicted values. The theoretical analysis was based on the
assumption that the unreactedness (the combustion inefficiency)
was represented by vaporized but unburned raw fuel. The predicted
heat release was therefore reduced directly with the unreacted-
ness. Experimentally, however, the unburned material which is
most significant in the combustion chamber under lean operating
conditions is not raw fuel, but rather carbon monoxide. In
order to correlate the CO data, we therefore assumed that the
CO levels would be equal to the predicted unreactedness. This
is clearly inconsistent. If we had refined our analysis to allow
for partially reacted material leaving the chamber (the fuel
converting into water vapor and carbon monoxide), then the
heat release used in the analysis at low efficiency points
would have been greater and the predicted emissions would have
been less as the blow out limits were approached. A revision
of the analysis to account for this partial reaction process
would be desirable, but it was not feasible within the
financial limitations of the present contract. In any case,
it seems clear that the CO emissions to be expected from the
Paxve burner will be less than those predicted by the present
theory. This gives us a method for obtaining CO emission
estimates which will be conservative for burner application
studies.
Figure 6 also shows the dramatic reduction in CO
emissions v/hich occur as the exhaust gases pass through the
vapor generator stack. This is undoubtedly a result of the
continued oxidation of the CO to CO? which occurs as the gases
cool off during their passage through the heat exchanger. The
rate of the CO oxidation reaction is still substantial at
temperatures over 2000°F. As the gases cool, there is time
for the combustion reaction to come closer to completion.
There is also time for CO which arises from the high temperature
equilibrium dissociation reaction to recombine as the equilibrium
shifts during cooling. The theoretical predictions are thus
even more conservative for a burner coupled to a heat exchanger.
VIII-8

-------
burner stability correlation
0.5
BURNER
1.0
BURNING 2.00-
LIMIT 1.00
BLOWOUT
STAB ILITY
1.5
	I	
PREDICTION
2.0
—|	
PARAMETER
2.5
INTR
3.0
H-
3.5
4.0
0- -+
M—
0. 5
1.0	1.5	2.0	2.5	3.0
BURNER STABILITY PREDICTION PARAMETER INTR
3.5
4.0

-------
Hi
O
0.001
0.01	o.l
INTENSITY PARAMETER INTD
1.0
10

-------
CORRELATION OF OXIDES OF NITROGEN DATA
WITH RECIPROCAL COMBUSTION TEMPERATURE
CORPELATION OF OXIDES OF UITPOGEN DATA
KFROSENF	HOT	BUPNER DATA
NOX CONCENTRATION NOB PPM
Fig.
VIII-3
Fi*.
vm-i*

-------
CORRELATION OF BURNER OXIDES OF NI TROGEN DATA
ta
o
i
x
E-
<
C3C
n .00
(X10) COPRELATING PARAMETER N0X/02
0,10 0.22 0.46 1.00 2.]S 4.6*J
r,	—H	—f	—	1	h-	1	
I
3.75j+
!
1
I
!
3.50-J-+
10.00
u:
t-
E-h
to
n
cc
c
o
.J
<
o
c
e:
p
t-H
o
UJ
a:
i
I
!
3.25j+
!
3.00j+
!
!
I
8
2.7r>j+
e
!
M TO
I •
t
\
2.r>o-*H-
1 /TEMP &¦•+«	
0.10
-h
+
0.??	0.^	1.°n	?.]f	H.6H
CX10) CORPELAT I NC- PARAMETER	NOX/P?
3 0.00

-------
0.02	0.0H	0.06	0.08	0.10
NOMINAL FUEL AIR RATIO FAN
0.12
O.M
0.16
Fig.
VIII-6

-------
PAXVE BURNER
CARBON MONOXIDE DATA
KEROSENE DATA
VOLUME 52.3 cu in
AIR TEMP OVER 250 °F
100
10
.0
.1
NOMINAL FUEL AIR RATIO FAN

-4#
¦4hv





\
~\—
r
1
V \
T
i
t , !
; • i



—
¦ V
X


-
| ;
i
tl{
	 1 ..
¦\h
A
-¦J-

.11
!:
i

]
.;U<
\

-------
PROGRAM PREDICT
7 PREDICT ; I ; 77/ \TI \INTD\LIT\CA ;CV-.CP-.TEIT
1]	NL+pPHIC
2]	!«-1
3 ] INTD14-IMTL+INTR+TBL1+EL+EFF+-TP* (pPHIC) p 0
4]	CO:P//-PffIff[J]
5]	n+TMCCl]
6]	f7C J]
7]	IHTDllII+INTDi-WADxCOtlSTlIl
8]	LIMIT
9]	IHTRiI]*-INTDHNT
10]	INTLII1--INT
11]	TBL\II~\*-TB
12]	ELlI~]*-E
13]	LIT+IIITDSINT
14]	•+( LIT = 0 ) /OUT
15]	BRN-.E+-1 - (l-E)*(.IHTD*It!T ) *1 - ( P'/ = 1 ) * ?
16]	CAL
17]	TEST+(1-INTD*IRT)
18]	¦~(( \TEST)>0 .01)/PRN
19]	OUT:EPFl I~\-LIT*E
20]	TPtI~}i-(T1]xLIT) + ( 1-LIT)*TI-*Z>0
21 ] I
22]	I-I +1
23]	-*(IiNL)/nO	Table
V VIIl-i
PROGRAM LIMIT
V LIMIT\I-,COUHT\INTM
[1]	COUNT+O
[2]	£«-0.3 + 0.05x>iM
[3]	ITT:COUNT<-COUllT+l
[4]	C\4L
[5]	INTM*-[/IUT
[6]	I-INT\INTU
[7]	-*( 1=1 )/0t/T
[8]	-+(COUNT = 3)/OUT
[9]	£-E[J-ll+CO.l*COUNT+l) *i10
[10]	-*ITT
[11]	OUT-.E+EII]
[12]	INTi-INTtll
[13]	TS+Mri]
Table
VIII-2

-------
1
R
1
1
1
1
1
1
1
2
1
0
1
0
1
1
2
2
1
2
2
2
0
0
1
1
1
COMPARISON OF PREDICTED AND EXPERIMENTAL BURNER
STABILITY
nin
VOL
AIR TEMP
AIR FLOV
EQUIV
INT DATA
HIT LIM
10.
IN*3
oF
LBSJHR
RA TIO
W/VP*2
W/VP*2
i
33.0
7 5
22.3
0.4678
0.3244
0.5487
2
33.0
80
22.2
0.5309
0.3229
1.5621
3
33.0
85
45.3
0.4822
0.6585
0.7412
4
33 . 0
85
45.3
0.4524
0.6585
0.4299
5
33.0
8 5
47.0
0.4650
0.6830
0.5447
6
3 3.0
8 a
61 . a
0.4501
0.8991
0.4276
7
33.0
90
61 . 7
0.4948
0.8974
0.9350
8
33.0
90
61.7
0.6326
0.8974
5.4903
9
33.0
90
61.7
0.5073
0 . 8974
1.1420
10
33.0
90
61.7
0.4697
0.897 4
0 .6079
11
33.0
90
61.7
0.5230
0 . 8974
1 .4478
12
33.0
90
61 . 7
0.4322
0.8974
0.2935
13
33.0
90
61.7
0.5386
0.897 4
1.8107
14
3 3.0
90
61.7
0.5073
0.897 4
1.1420
15
33.0
90
61.7
0.5089
0 . 897 4
1.1695
16
33.0
90
78.1
0.5689
1.1362
2.6981
17
33.0
90
78.1
0.5417
1.1362
1.8890
18
33.0
90
78.1
0.5541
1.1362
2.2311
19
33.0
91
7 8.0
0.6115
1.1352
4.43 20
20
33.0
92
78.0
0.5377
1.1341
1 .8009
21
33.0
92
7 8.0
0.4609
1 .1341
0.5216
22
33.0
93
77.9
0.4960
1.1331
0.9661
23
33 .0
93
77.9
0.5209
1.1331
1.4192
24
33.0
93
77.9
0.5382
1 .1331
1.8199
25
33.0
93
77.9
0.5631
1 . 1331
2.5330
BURN
2	
STABLE OPERATION

UNITS OF
INT ARE
BURN
1	
STABILITY LIMIT



BURN
0	
BURNER GOES
OUT




-------
COMPAFISON OF PREDICTED AND
BUN	VOL AIR TEMP AIR FLOW
NO.	IN*3	OF	LBS/HR
26
33.0
90
103.3
27
33.0
93
103.1
28
33.0
96
102.8
29
33.0
97
102.7
30
33.0
97
102.7
31
33.0
102
120.9
32
33.0
105
120.6
33
33.0
108
120 . 2
3 4
33.0
95
121.6
35
33 . 0
102
120.9
36
33.0
112
153.8
37
33.0
117
153 .2
38
33.0
120
152.8
39
33 . 0
120
152.8
40
33.0
120
152.8
i+l
33.0
124
152.3
<+2
33.0
91
103 .2
43
33.0
93
103.1
44
33.0
87
41.8
45
33.0
85
41 . 9
46
33.0
84
41.9
47
33.0
82
42.0
48
33.0
82
42.1
49
33.0
80
rl
CM
50
33.0
80
42.1
BURN = 2 	 STABLE OPERATION
BURN = 1	STABILITY LI.'HT
BURN = 0 	 BURNER GOES OUT
<
h-l Q,
K. tr
• f-
¦c- (B
EXPERIMENTAL BURNER STABILITY DATA
PAGE 2

EQUIV
INT DATA
INT LIM
INT RAT
RATIO
W/VP*2
y/vp*2
ID/IL
0
. 6937
1.5030
9.5186
0.1578
0
. 5794
1.4989
3 . 0973
0.4835
0
. 5227
1 .4949
1.4727
1.0140
0
.5118
1.4935
1.2588
1.1852
0
. 5231
1.4935
1 .4887
1.0022
0
. 5930
1.7583
3.7383
0.4699
0
. 5465
1.7537
2.1285
0.8231
0
.5158
1.7490
1.3937
1.2537
0
. 5099
1.7694
1.2117
1.4587
0
.5190
1.7583
1.4273
1 . 2307
0
. 5878
2.2377
3.6370
0.6147
0
.5613
2.2280
2.6910
0.8272
0
. 5451
2.2222
2 . 2041
1.0073
0
. 5299
2.2222
1.7910
1.2396
0
.5135
2.2222
1.4105
1.5739
0
. 5470
2.2146
2.2941
0.96 54
0
. 7 565
1 . 5016
15.0328
0.0998
0
. 5222
1 .4989
1.447 4
1.0345
0
. 7 529
0 . 60 8 5
14.5467
0 .0418
0
.6916
0.6096
9.2316
0.0660
0
.6132
0.6096
4.4163
0.1379
0
. 5333
0.6113
1.6303
0.3745
0
.4910
0.6124
0.8497
0.7199
0
.4497
0.6124
0.3988
1.5337
0
. 4997
0.6124
0.9710
0 .6300


UNITS OF
INT ARE
LBS/SEC
BURN
2
2
0
0
1
2
2
1
0
1
2
2
2
0
*1
1
2
1
2
2
2
2
2
0
1

-------
COMPARISON OF PREDICTED AND EXPERIMENTAL BURNER STABILITY DATA
PAGE 3
RUN
NO.
VOL
IN* 3
AIR TEMP
OF
AIR FLOW
LBS/HR
EQUIV
RA TIO
HIT DATA
U/VP*2
INT LIM
V/VP*2
INT RAT
ID/IL
BURN
51
33.0
83
2 5.4
0.6097
0 .3688
4.2404
0.0869
52
33.0
82
25.4
0.5406
0 .3691
1.8068
0 .2041
53
33.0
82
25.4
0.464 5
0 . 3691
0.5329
0.6919
54
33.0
83
25.4
0.4510
0.3688
0 . 4144
0 .8888
55
"l . 0
83
42.0
3.17 53
~1 . 0000
"1.0000
"1 .0000
56
~1 . 0
81
42.1
2.6412
"l . 0000
"l.0000
~1.0000
57
~1 . 0
82
42.0
2 . 8966
"l . 0000
"l .0000
~1.0000
58
1 . 0
8 5
41.9
2.6510
"l . 0000
"l.0000
"1.0000
59
"l . 0
8 5
41.9
3.1812
"l.0000
"l.0000
"l .0000
60
~1 . 0
90
41.7
1.94 52
1.0000
"1.0000
"1.0000
61
~1 . 0
90
41.7
1.9452
"l . 0000
"l.0000
"l . 0000
62
~1 . 0
85
41.9
2.6510
"l .0000
"l.0000
"l.0000
63
~1 . 0
B 7
41.8
3.1870
~1.0000
"l . 0000
"1.0000
64
~1 .0
85
41.9
2.6510
"l . 0000
"l . 0000
"l .0000
65
"l . 0
88
41.8
2.7692
~1.0000
"1 . 0000
"1.0000
66
"l .0
85
41.9
2.8585
"l.0000
"l.0000
~1 .0000
67
~1 .0
85
41.9
2.7616
"l.0000
"l .0000
"1.0000
68
~1 .0
78
25.5
2.5262
~1 . 0000
~1 .0000
"l .0000
69
~1 .0
82
24 . 7
2.3985
"l.0000
"l.0000
~1.0000
70
1.0
85
25.3
2.9167
"l . 0000
"l.0000
~1.0000
71
~1 . 0
86
25.3
3.0646
"l . 0000
"l . 0000
"1.0000
72
~1 .0
90
61.7
2.3863
~1.0000
"l.0000
"1.0000
73
"1 .0
93
61.5
2 . 4493
"l.0000
~1 . 0000
"1.0000
74
33.0
86
25.3
0.5044
0.3677
1.0731
0.3423
75
33.0
8 5
25.3
0 .6219
0.3681
4.8599
0 .07 57
BURN = 2 	 STABLE OPERATION	UNITS OF INT ARE LBS/SEC FT*3 ATM*2
BURN = 1 	 STABILITY LIMIT
BURN = 0 	 BURNER GOES OUT
c -3
hH Q)
w er
M I—*
I 
-------
COMPARISON OF PREDICTED AND EXPERIMENTAL BURNER STABILITY DATA	PACE 4
RUN
NO.
VOL
IN*3
AIR TEMP
OF
AIR FLOW
LBS/HR
EQUIV
RATIO
INT DATA
W/VP*2
INT LIM
W/VP*2
INT RAT
ID/IL
BURN
76
33.0
85
25.3
0.8339
0.3681
22.9747
0.0160
2
77
33.0
85
25.3
0 . 9960
0.3681
41.5938
0.0088
2
78
33.0
90
61.7
0.5571
0.8974
2.3338
0.3845
1
79
33.0
92
61.6
0.6869
0.8958
9.0511
0.0989
2
80
33.0
90
61 . 7
0 . 9927
0.8974
41.6033
0.0216
2
81
"l .0
92
61 . 6
1.4 432
"l.0000
"l.0000
"l.0000
~1
82
33.0
103
102.1
0.5050
1.4855
1 .1611
1 .2781
2
83
33.0
105
101.9
0.7666
1.4829
16.5407
0 .0896
2
84
33.0
100
102.1
0.7464
1.4895
14.3647
0 .1036
2
85
33.0
103
102.1
0.9178
1.4855
33.8123
0 .0439
2
86
"l .0
105
101.9
1 .2812
~1.0000
"1.0000
"l.0000
~1
87
33.0
300
59 . 3
0.5391
0.8628
3.8187
0.2259
2
88
33.0
300
59 . 3
0 . 6869
0.8628
15.3308
0.0563
2
89
"1.0
300
59.3
1.1075
~1.0000
"1.0000
"1.0000
"1
90
33.0
300
59 . 6
0.5770
0.8665
5.8483
0.1482
2
91
33.0
310
59 . 3
0.8222
0.8628
34.3220
0.0251
2
92
"1.0
300
59 . 3
1.1075
"l . 0000
"1 . 0000
~1.0000
~1
93
52.3
250
165.3
0.4863
1.5174
1.5676
0.9680
2
94
52.3
250
165.3
0.6478
1.5174
9 .9126
0 .1531
2
95
52.3
250
165.3
0 . 8299
1.5174
31.5282
0.0481
2
96
52.3
250
132.9
0.9865
1.2193
54.2496
0 .0225
2
97
52.3
95
55.9
0.4881
0.5126
0.8554
0 . 5986
2
98
52.3
95
55.9
0.6175
0.5126
4.7821
0 .1071
2
99
52.3
95
55.9
0.8087
0.5126
20 . 6232
0.0248
2
100
52.3
100
55.6
0.9984
0.5103
42.9648
0.0119
2
BURN = 2 	 STABLE OPERATION	UNITS OF INT ARE LBS/SEC FT*3 ATM*2
BURN = 1 	 STABILITY LIMIT
BURN = 0 	 BURNER GOES OUT
< -J
M&
l-H f_

-------
5
R
1
2
2
2
2
1
0
2
2
2
2
2
2
2
1
2
2
0
2
2
2
0
2
1
1
COMPARISON OF PREDICTED AND EXPERIMENTAL BURNER STABILITY
RUN
VOL
AIP TE11P AIR FLOW
EQUIV
INT DATA
INT LIM
NO.
IN*3
OF
LBS/HR
RATIO
W/VP*1
IH VP* 2
101
~1 .0
102
55.5
1.1925
0.0000
0.0000
102
52.3
102
89.0
0.7520
0.8165
14.9787
103
52.3
4 0 0
50 .0
0.8359
0.4593
43.4487
104
52.3
400
50 . 0
0.5406
0.4593
5.4104
105
52.3
84
49.9
0.9250
0.4581
33.4897
106
52.3
8 5
49.9
0.4888
0.4581
0 . 8290
107
52.3
85
49.9
0.3925
0.4581
0.1159
108
52 . 3
85
49.9
0.4448
0.4576
0.3703
109
52.3
85
49 . 9
0.5563
0.4576
2.2581
110
52.3
400
50 . 0
0.4039
0 .4593
0 .7676
111
52.3
400
50 . 0
0 .3915
0 .4 593
0.6090
112
52.3
400
33.2
0.46B8
0.3048
2.1848
113
52.3
400
33.2
0 . 3B91
0.3048
0.5814
114
52.3
400
33.2
0.3805
0.3043
0.4912
115
52.3
400
33.1
0.3883
0.3040
0.5727
116
52.3
400
82.7
0.5375
0.7587
5.2248
117
52.3
400
82.8
0.4312
0.7601
1.2303
118
52.3
400
82.8
0.3861
0.7601
0.5487
119
52.3
400
126 . 8
0 . 4672
1 .1634
2.1364
120
52.3
400
127.0
0 .4279
1.1654
1.1641
121
52.3
400
126.8
0 .4469
1.1634
1.5783
122
52.3
400
126 . 8
0.4010
1.1634
0.7272
123
52.3
74
50 . 5
0.7951
0.4634
18.2890
124
52 . 3
80
49 . 5
0.4743
0 . 4542
0.6308
125
52.3
85
47.6
0.4930
0.4369
0.8895
BURN =
: 2	
STABLE OPERATION

UNITS OF
INT ARE
BURN ¦-
: 1	
STABILITY
LIMIT



BURN =0 	 BURNER GOES OUT

-------
COMPARISON OF PBEDICTED AND EXPERIMENTAL BURNER STABILITY DATA	PACE 6
RUN	VOL AIR TEMP
NO.	IN*3	OF
126	52.3	95
127	52.3	95
128	52.3	90
129	52.3	80
130	52.3	80
131	52.3	80
132	"1.0	80
133	52.3	80
134	52.3	80
135	52.3	80
136	52.3	88
137	52.3	90
138	52.3	95
139	52.3	95
140	1.0	95
141	"1.0	95
142	52.3	95
143	"1.0	95
144	"1.0	95
145	52.3	95
146	52.3	95
147	52.3	95
148	52.3	95
149	"1.0	95
150	52.3	95
AIR FLOW	EQUIV	INT DATA
LBS/11R	RATIO	VI VP* 2
127.3	0.5049	1.1684
119.6	0.5687	1.0976
57.8	0.5411	0.6226
86.2	0 . 5061	0 .7910
82.8	0.5822	0 . 7604
66.7	0.6869	0.6124
44.5	1.0553	0.0000
B3.4	0.5628	0.7655
66.7	0.7036	0.6124
72.3	0.4347	0.6634
72.3	0.5318	0.6634
66.7	0.7036	0.6124
96.6	0.6404	0.8867
95.0	0.7260	0.8715
91.1	1.0266	0.0000
77.3	1.2099	0.0000
77.8	0.5389	0.7145
77.8	1.2013	0.0000
91.1	1 . 0266	0 . 0000
94.4	0.7302	0.8665
96.6	0.5433	0.8867
98.3	0.4702	0.9019
90.5	0.7614	0.8310
53.6	1.2873	0.0000
56. 9	0 . 4388	0 . 5219
INT LIM	INT RAT	BURN
W/VP*2	ID/IL
1.1230	1.0404	0
2.7371	0.4006	1
1.8745	0.3318	2
1.0758	0.7344	2
3.0707	0.2474	2
8.7686	0.0698	2
0.0000	0.0000	"1
2.4145	0.3167	2
10.0334	0.0610	2
0.2940	2.2538	0
1.6320	0.4061	2
10.2913	0.0594	2
6.0209	0.1471	2
12.3178	0.0707	2
0.0000	0.0000	"l
0.0000	0.0000	"l
1.8518	0.3854	2
0.0000	0.0000	"1
0.0000	0.0000	~1
12.6988	0.0682	2
1.9654	0.4507	2
0.6264	1.4383	1
15.6459	0.0531	2
0.0000	0.0000	"1
0.3447	1.5126	2
BURN = 2 	 STABLE OPERATION
BURN = 1 	 STABILITY LIMIT
BURN - 0 	 BURNER GOES OUT
UNITS OF INT ARE LBS/SEC FT*3 ATM*2
<
~H	-3
W	fll
n	cr
I	y-
ro	(1>

-------
COMPARISON OF PREDICTED AND EXPERIMENTAL BURNER STABILITY DATA	PAGE 7
RUN
VOL
AIR TEMP AIR FLOW
EQUIV
INT DATA
INT LIM
INT RAT
BURl
NO.
IN*3
OF
LBS/11R
RATIO
W/VP*2
W/VP*2
ID/IL

1 51
52 . 3
400
84.0
0 .7066
0.7706
22.2613
0 .0346
2
1 52
52 . 3
400
79.5
0 .7461
0. 7297
28.1965
0.0259
2
153
52 . 3
360
111.2
0.4703
1 .0206
1.9112
0.5341
1
154
52.3
355
55.6
0.7013
0.5103
19 . 3644
0.0263
2
155
52 . 3
270
98.9
0.5043
0.9081
2.1939
0.4136
2
156
52.3
260
103.3
0.4292
0 .9484
0.6331
1.4983
0
157
52.3
260
100.0
0.5856
0 .9182
5.6586
0.1622
1
158
52.3
260
1 00 . 6
0.6375
0.9232
9 . 3190
0.0990
2
1 59
52.3
265
97 . 3
0 .7598
0.8929
22.6992
0.0393
2
160
~1 .0
270
90. 7
1.2226
0 .0000
0 . 0000
0 .0000
"l
161
~1 . 0
275
89 . 6
1 .2418
0 .0000
0.0000
0.0000
~1
162
~1 . 0
275
87.9
1.5592
0 .0000-
0 . 0000
0 .0000
"1
163
52.3
360
22.0
0 . 5464
0.2018
5 .0677
0.0398
1
161
52.3
370
22.0
0.7565
0 . 2018
27 . 9911
0.0072
2
165
52.3
375
30.8
0.7925
0.2825
34.0641
0.0083
2
166
52.3
36 5
44. 5
0.6641
0.4086
15.2380
0.0268
2
167
52 . 3
370
44.0
0.8406
0.4036
41 .7926
0.0097
2
168
52 . 3
370
41 . 2
0.9728
0.3784
64 . 4761
0.0059
2
169
52 . 3
345
55.0
0.7296
0.5045
22.6681
0.0222
2
170
52 . 3
325
69 . B
0.5771
0 .6407
6.3214
0.1013
2
171
52 . 3
305
78.1
0.5162
0.7164
2.9222
0.2450
2
172
52 . 3
275
97.8
0.4118
0.8980
0.4885
1.8383
0
173
"l . 0
"l
~1 .0
~1 . 0000
0.0000
0 . 0000
0.0000
"l
174
52.3
400
96.6
0.6507
0.8864
15.0295
0.0589
2
175
52 . 3
405
93 . 3
0.7925
0.8561
36.1910
0.0236
2
BURN =
2	
STABLE OPERATION

UNITS OF
INT ARE
LBS/SEC FT*3
ATM
BURN =
1	
STABILITY
LIMIT





BURN s 0 	 BURNER GOES OUT
<
Kg-
I f—
to CD

-------
COMPARISON OF PREDICTED AND EXPERIMENTAL BURNER STABILITY DATA	PAGE 8
RUN
VOL
AIR TEMP
AIR FLOW
EQUIV
INT DATA
INT LIM
INT RAT
BURl
NO.
IN* 3
OF
LBS/IIR
RATIO
W/VP*2
W/VP*2
ID/IL

176
52.3
40 5
93.1
0.8733
0.8547
50.6320
0 .0169
2
177
52.3
105
98.6
0.4239
0.9049
1.1124
0.8135
2
178
52.3
4 0 5
95.9
0.5413
0.8806
5.5355
0.1591
2
179
52.3
350
133.7
0.8388
1.2275
39.9028
0.0307
2
180
52.3
345
134.6
0 .7140
1.2354
20.5484
0.0601
2
181
52.3
340
136.1
0 . 6205
1.2493
9.9945
0.1249
2
182
52.3
340
138.8
0.5458
1.2743
4.7161
0.2700
2
183
52.3
335
139.0
0.4894
1.2754
2.2792
0 . 5596
2
184
52.3
334
141.7
0 .4435
1.3004
1 .1318
1.1491
1
185
52.3
335
139.0
0.5014
1.27 54
2.6829
0 .4754
2
186
52.3
335
139.0
0.5631
1 . 2754
5.6213
0 . 2269
2
1 87
"l .0
~1
~1 . 0
~1.0000
0.0000
0.0000
0.0000
"l
188
52.3
430
77.9
0.6830
0.7145
20.4438
0 .0349
2
1 89
52.3
430
79.0
0.5532
0.7252
6.8132
0.1065
2
190
52.3
430
79.0
0.4885
0.7252
3.2004
0 . 2266
1
191
52.3
400
66.9
0.5766
0.6144
7.8786
0.0780
2
192
52.3
405
56.0
0.6896
0.5137
20.1414
0.0255
2
193
"1 .0
410
51. 6
1.2578
0.0000
0 .0000
0.0000
"1
1 94
52.3
410
113.0
0.5463
1 . 0374
5.9429
0.1746
"1
195
52.3
410
113.6
0.4858
1.0425
2.8719
0 .3630
"l
196
52.3
410
115.2
0.4920
1 . 0576
3.1155
0.3395
"l
197
52.3
430
101.4
0.5304
0.9309
5.3194
0.1750
~1
198
52.3
360
41.2
0.6579
0.3780
14.3547
0.0263
"l
199
52.3
340
38.5
0.9818
0 .3 531
62.6358
0.0056
"l
200
52.3
410
38.5
0.6616
0.3531
16.7434
0.0211
"l
BURN :
= 2	
STABLE OPERATION

UNITS OF
INT ARE
LBS/SEC FT*3
ATM
BURN -¦
: 1	
STABILITY LIMIT





BURN = 0 	 BURNER GOES OUT
<
<	Qj
h-»	cr
I	H
M	ft
O

-------
9
R
2
2
2
1
2
2
2
2
2
2
2
2
2
2
2
COMPARISON OF PREDICTED AND EXPERIMENTAL BURNER STABILITY
RUN
VOL
AIR TEMP
AIR FLOW
EQUIV
INT DATA
INT LIM
NO.
IN* 3
OF
LBS/HR
RA TIO
W/VP*2
W/VP*2
201
52 .3
8 6
83 . 2
0 .8221
0.7634
21.7050
202
52.3
82
102.9
0 .7187
0.9441
11.3179
203
52.3
110
148 .0
0.6216
1.3581
5.2160
204
"1 . 0
"1
~1 .0
~1.0000
0 .0000
0 . 0000
205
52.3
313
76 . 8
0 .6978
0.7049
17.1035
206
52.3
253
167.8
0 .6498
1.5402
10.1619
207
52.3
245
1 81 . 5
0 . 5601
1 . 6655
4.0754
208
66.5
70
51 . 5
0.8177
0.3718
20.5144
209
66 . 5
70
49 . 9
0 .9883
0.3598
39.6679
210
66.5
70
49.9
0.6666
0.3598
7.1656
211
"l . 0
70
49.9
1.0852
0 .0000
0 . 0000
212
66.5
70
102.9
0.6368
0.7425
5.4032
'13
66 . 5
70
99.7
0.7865
0.7199
17 . 2529
^14
"1 . 0
70
96.6
1.0999
0.0000
0.0000
215
66.5
70
49 .9
0.4845
0.3598
0 . 7215
218
66.5
70
48.2
0.6615
0 . 3479
6 . 8425
217
52.3
80
83.8
0.6044
0.7689
3.9655
218
66 . 5
100
91 . 5
0.6187
0.6606
4.9179
219
66 . 5
110
95. 3
0.7845
0.6B81
18.6383
220
52.3
100
86 . 2
0.5854
0.7910
3.403 6
221
52.3
100
116.8
0.5761
1.0717
3 . 0495
222
52.3
100
131.8
0.5105
1.2094
1.2482
223
52.3
100
108.4
0.6205
0.9951
5 . 0047
224
52.3
100
168.5
0.5562
1.5462
2.3779
225
52.3
100
151.8
0.6173
1.3932
4.8472
BURN = 2 	 STABLE OPERATION
BURN =1 	 STABILITY LIMIT
BURN = 0 	 BURNER GOES OUT
UNITS OF
INT ARE

-------
COMPARISON or PREDICTED and experimental burner stability data
PAGE 10
RUN	VOL	AIR TEMP
NO.	IN*3	OF
226	52.3	100
227	52.3	76
228	52.3	82
229	52.3	85
230	52.3	433
231	52 . 3	4 20
232	52.3	410
233	52.3	407
234	52.3	90
235	52.3	92
236	52.3	89
237	52.3	82
238	66.5	79
239	66.5	80
240	66.5	82
241	66.5	82
242	66.5	78
243	66.5	80
244	66.5	79
245	66.5	80
246	66.5	80
247	"1.0	75
248	66.5	85
249	66.5	75
250	66.5	85
AIR FLOW	EQUIV	INT DATA
LDS/HR	RATIO	W/VP*2
177.9	0.5266	1.6330
55.8	0.5796	0.5121
58.4	0.5761	0.5358
57.6	0.5773	0.5288
57.6	0.5779	0.5284
60 .8	0.5473	0.5579
90.9	0.5203	0.8346
90.2	0.6172	0.8281
76.9	0.6725	0.7062
77.3	0.5930	0.7094
77.7	0.6003	0.7132
38.4	0.6353	0.3524
49.1	0.4880	0.3544
48 .9	0 . 5391	0 . 3531
48.9	0.6563	0.3528
48.8	0.7266	0.3522
49.2	0.7617	0.3554
48.8	0.8945	0.3522
98.1	0.5043	0.7082
97.6	0.5994	0.7043
60.5	0.6393	0.4364
"l.O	0.4609	0.0000
47.2	0.5625	0.3404
49.2	0.8476	0.3554
48.8	0.8555	0.3522
INT LIM	INT RAT	BURN
W/VP*2	ID/IL
1.5866	1.0293	1
2.9354	0.1743	2
2.8714	0.1864	2
2.9431	0.1795	2
8.7961	0.0601	2
6.2023	0.0900	2
4.4305	0.1884	2
11.6716	0.0709	2
7.9725	0.0885	2
3.6204	0.1957	2
3.8997	0.1829	2
5.5154	0.0638	2
0.7973	0.4445	2
1.7569	0.2010	2
6.7467	0.0523	2
11.9947	0.0294	2
15.0894	0.0236	2
29.7059	0.0119	2
1.0409	0.6796	2
3.7461	0.1878	2
5.7049	0.0765	"l
0.0000	0.0000	~1
2.4459	0.1390	"l
24.0592	0.0148	"l
25.4455	0.0138	~1
BURN = 2 	 STABLE OPERATION	UNITS OF INT ARE LBS/SEC FT*3 ATM*2
BURN = 1	STABILITY LItllT
BURN =0 	 BURNER GOES OUT
<
i-"	-J
M	Qj
i-i	cr
i	>-•
*-•	IS
IO

-------
COMPARISON OF PREDICTED AND EXPERIMENTAL BURNER STABILITY DATA	PAGE 11
RUN
POL
AIR TEMP
AIR FLOW
EQUIV
INT DATA
INT LIM
INT RAT
BURN
NO.
IN+3
OF
LBS/HR
RATIO
W/VP*2
W/VP*2
ID/IL

251
66 . 5
8 S
100 . 8
0.4983
0 . 7278
0.9702
0.7502
"1
252
66 . 5
90
100.4
0 . 5006
0.7245
1.0273
0 . 7052
~1
253
66 . 5
92
97 . 0
0 . 5772
0 . 6998
3.0094
0.2326
"l
254
66.5
93
100 .1
0.6412
0.7225
6.0290
0.1197
"1
255
52.3
93
90.0
0 .6063
0 . 8260
4 .2127
0.1959
2
256
66 . 5
8 5
97.6
0.7188
0 .7043
11 .4080
0.0617
2
257
66 . 5
85
97.6
0.7852
0 .7043
17.6921
0.0398
2
258
66.5
85
97.5
0.9004
0.7037
30.6648
0 .0229
2
259
66 . 5
85
97.4
1 . 0000
0 .7030
42.0062
0.0167
2
260
66. 5
8 5
97 . 3
0.9609
0.7024
37.6931
0.0186
2
261
~1 .0
89
97 . 2
1.1875
0.0000
0 .0000
0 . 0000
~1
262
66. 5
90
144 . 4
0.5167
1.0421
1.3178
0 .7900
2
263
66.5
97
143.5
0.5957
1.0355
3.7940
0.2727
2
264
66.5
8 5
48.8
0 .4951
0.3522
0.9199
0 . 3824
2
265
66. 5
85
48.8
0.5820
0.3522
3.1141
0.1130
2
266
66.5
96
143.0
0 . 5677
1 .0319
2.7115
0.3802
2
267
66 . 5
96
97 . 8
0.5045
0.7063
1.1193
0 .6304
2
268
66 . 5
96
97.2
0.5817
0.7018
3.2131
0 .2182
2
269
66.5
96
97 . 2
0.6582
0 .7018
7.1329
0.0983
2
270
66 . 5
96
97 . 1
0 .6860
0.7011
9 .0735
0.0772
2
271
66 . 5
100
144. 1
0 . 5576
1 .0402
2.4183
0.4297
2
272
66. 5
100
141 . 7
0 .6618
1 .0231
7.4548
0.1371
2
273
66 . 5
100
143. 1
0 .6745
1.0328
8 . 3279
0.1239
2
274
66 . 5
100
143.5
0.7565
1.0355
15.3389
0.0675
2
275
66.5
95
143.7
0.8254
1 .0374
22.4802
0.0461
2
BURN - 2 	 STABLE OPERATION
BURN = 1 	 STABILITY LIMIT
BURN = 0 	 BURNER GOES OUT
UNITS OF INT ARE LBS/SEC FT*3 ATM*2
<
hH H
»—1 Qj
KX
I I—'
h-
u

-------
COMPARISON OF PREDICTED AND EXPERIMENTAL BURNER STABILITY DATA	PAGE 12
RUN	VOL	AIR TEflP
NO.	IN*3	OF
276	"1.0	"1
277	"1.0	~1
278	"l.O	"l
279	52.3	109
280	52.3	110
281	52.3	110
282	52.3	94
283	52.3	100
284	52.3	105
285	52.3	116
286	66.5	82
287	66.5	92
288	66.5	90
289	66.5	95
290	66.5	99
291	"l.O	73
292	"l.O	75
293	"l.O	82
294	"l.O	90
295	"l.O	72
296	"l.O	86
297	"l.O	92
298	"l.O	95
299	"l.O	96
300	"l.O	85
AIR FLOW	EQUIV	INT DATA
LBS/UR	RATIO	W/VP*2
130.0	"1.0000	0.0000
140.0	"1.0000	0.0000
"l.O	"1.0000	0.0000
91.0	0.5791	0.8353
90.9	0.6808	0.8346
111.3	0.5932	1.0217
4 C . 4	0 . 6072	0 . 4258
47.3	0.6024	0.4338
46.5	0.4239	0.4268
136.5	0.4873	1.2529
58.8	0.5010	0.4246
144.0	0.5260	1.0394
144.3	0.5654	1.0413
143.6	0.6016	1.0366
142.9	0.7193	1.0318
66.0	1.3984	0.0000
49.1	1.5625	0 .0000
47.1	1.3594	0.0000
51.1	1.2578	0. 0000
49.2	1.3203	0. 0000
48 . 6	1 . 3438	0 .0000
48.3	1.4141	0. 0000
48.2	1.2813	0. 0000
87.2	1.3750	0.0000
49.3	1.1719	0.0000
INT LIM	INT RAT	BURN
W/VP* 2	ID/IL
0.0000	0.0000	"1
0.0000	0.0000	"l
0.0000	0.0000	"l
3.2549	0.2564	2
9.0216	0.0924	2
3.8430	0.2656	2
4.2688	0.0997	2
4.1303	0.1050	2
0.2659	1.6031	2
0.9210	1.3590	2
1.0009	0.4237	2
1.5258	0.6812	2
2.5809	0.4030	2
4.0261	0.2572	2
11.8468	0.0870	2
0.0000	0.0000
0.0000	0.0000
0.0000	0.0000
0.0000	0.0000
0.0000	0.0000
0.0000	0.0000
0.0000	0.0000
0.0000	0.0000
0.0000	0.0000
0.0000	0.0000
BURN =2 	 STABLE OPERATION	UNITS OF INT ARE LBS/SEC FT*3 ATM*2
BURN =1 	 STABILITY LIMIT
BURN = 0 	 BURNER COES OUT
H
0)

-------
COMPARISON OF PREDICTED AND EXPERIMENTAL BURNER STABILITY DATA	PAGE 13
RUN
NO.
VOL
IN*3
AIR TEMP
OF
AIR FLOW
LBS/HR
EQUIV
RATIO
INT DATA
V/VP*2
INT LIM
W/VP*7
INT RAT
ID/IL
BURN
301
1 . 0
88
48.5
1.2188
0.0000
0.0000
0 .0000
302
~1 . 0
92
96.9
1.4219
0.0000
0 . 0000
0 .0000
303
~1 . 0
74
98.5
1 .2813
0.0000
0 .0000
0 .0000
301
1 . 0
B 7
141.5
1 . 3438
0.0000
0 . 0000
0 .0000
305
~1 . 0
100
124.2
1 . 3750
0.0000
0 . 0000
0.0000
306
"l .0
100
139.8
1.1719
0.0000
0 . 0000
0 .0000
307
~1 .0
104
94.3
1.1953
0.0000
0 .0000
0.0000
308
52.3
90
46.6
0.5469
0.4279
2.0293
0.2109
309
52.3
98
47.5
0.5052
0.4362
1.1405
0.3820
310
52.3
94
48.4
0.3436
0.4443
0 .0318
13.9791
311
52.3
385
49.4
0.5886
0.4533
8 .4696
0.0535
312
52.3
388
48.3
0 . 5098
0 . 4432
3.6137
0.1226
313
52.3
395
47.2
0 . 3945
0.4331
0.6289
0.6887
314
52.3
340
88.2
0.5392
0.8094
4.3720
0.1851
315
52.3
364
93.1
0.4726
0 .8547
2.00 56
0 .4261
316
52.3
400
90.5
0.5043
0.8310
3.5181
0.2362
317
52.3
435
90.0
0.4059
0.8259
0 .9363
0.8821
318
52.3
350
137.1
0 .4935
1.2582
2.5485
0.4937
319
52.3
350
137.2
0.4150
1.2589
0 .7428
1.6949
320
52 . 3
351
134.5
0.6484
1.2340
13.0187
0.0948
321
52.3
1 40
89.8
0.5978
0.8238
4.4449
0.1852
322
52.3
115
90.0
0.4857
0.8259
0.8930
0.9240
323
52.3
105
90 .0
0 .4477
0.8259
0.4324
1.9101
3 2 4
52. 3
95
135.5
0.5847
1 . 2436
3.3180
0.3744
325
52.3
105
137.3
0.5181
1.2604
1.4278
0.8827
BURN =
2	
STABLE OPERATION

UNITS OF
INT ARE
LBS/SEC
BURN =
1	
STABILITY
LIMIT




BURN =0 	 BURNER GOES OUT
< -3
M Qj
m tr
t—
I ft

-------
COMPARISON OF PREDICTED AND EXPERIMENTAL BURNER STABILITY DATA
PAGE 1
run
VOL
AIR TEMP
AIR FLOW
EQUIV
INT DATA
INT LIM
INT RAT
BURI
so.
IN* 3
OF
LBS/HR
RA TIO
U/VP*2
U/VP*2
ID/IL

326
52 . 3
110
138.7
0 .4796
1
. 2732
0.7905
1.6107
1
327
52.3
412
48.2
0 . 6017
0
.4420
10.3270
0.0428
2
328
52.3
39 5
47.3
0.4641
0
.4338
2.0023
0.2166
2
329
52.3
390
47.1
0.3821
0
.4318
0.4819
0.8961
2
330
52.3
407
90.3
0.5955
0
. 8288
9.6208
0.0862
2
331
52.3
400
92.7
0 .4834
0
.8511
2.6750
0.3182
2
332
52 . 3
435
89 . 7
0.4059
0
. 8231
0.9363
0 .8791
1
333
52.3
392
135.2
0.4401
1
. 241 1
1.3715
0.9050
2
331
52.3
395
140.3
0.3843
1
. 2877
0.5173
2.4893
1
335
52.3
92
91.3
0.5556
0
. 8383
2.2910
0.3655
2
336
52.3
90
92.6
0 .4667
0
. 849B
0.5751
1.4759
1
337
52.3
90
48.5
0.5434
0
. 4447
1.9337
0.2297
2
338
52.3
90
49.6
0.5899
0
.4548
3.4713
0 .1309
2
339
52.3
1 00
136.7
0.5685
1
.2547
2.7770
0.4514
2
340
52.3
395
138.3
0.4997
1
. 2691
3.2593
0 .3894
2
341
52.3
1 50
92.4
0.5245
0
. 8483
1.8563
0.4566
2
342
52.3
90
93.3
0.5472
0
.8563
2.0349
0.4204
2
343
52.3
102
91.3
0.5273
0
.8381
1.6153
0.5189
2
344
52.3
1 00
91.9
0.4688
0
. 8439
0.6253
1 .3496
2
345
52.3
102
135.3
0.5445
1
. 2414
2.0530
0.6047
2
346
52.3
105
138.2
0 .4898
1
. 2682
0.9186
1 .3806
2
347
52.3
105
134.5
0.6523
1
. 2343
6.9392
0.1779
2
348
52.3
107
90 .1
0.6022
0
. 8274
4.2116
0.1963
2
349
52.3
100
91.9
0.5137
0
. 8431
1.3105
0 .6427
2
3 50
52.3
101
136.4
0.5697
1
.2514
2.8268
0.4423
2
BURN ¦¦
: 2	
STABLE OPERATION

UNITS OF
INT ARE
LBS/SEC FT*3
ATM'
BURN -
: 1	
STABILITY LIMIT






BURN = 0 	 BURNER GOES OUT
<
M
hH QJ
*-* cr
i »—
£ ®

-------
COMPARISON OF PREDICTED AND EXPERIMENTAL BURNER STABILITY DATA	PACE 15
RUN
NO.
VOL
IN* 3
AIR TEMP
OP
AIR FLOW
LBS/HR
EQUIV
RATIO
INT DATA
UIVP*2
INT LIM
f// VP* 2
INT RAT
ID/IL
BURN
351
52.3
105
92.4
0.6311
0.8484
5.6528
0 .1500
2
352
52.3
110
93.1
0.5036
0.8548
1 .1679
0.7312
1
353
52.3
120
137.7
0.6311
1.2636
5 . 9076
0 .2137
2
3 54
52.3
430
81 .4
0.5379
0.7471
5.7838
0.1292
2
355
52.3
438
82.0
0.4309
0.7528
1.4414
0.5223
2
356
52.3
4 50
43.8
0.5570
0 .4018
7. 5311
0 .0534
2
357
52.3
140
180.5
0.5053
1.6565
1 . 3504
1 .2255
2
356
52.3
110
46.8
0 .4673
0 .4299
0.6363
0 .6750
2
359
52.3
120
138.2
0.5022
1.2686
1 .1888
1 .0661
2
360
52.3
380
92.6
0.5322
0.8498
4 . 6028
0 .1846
2
361
52.3
400
90.9
0.4673
0.8339
2.1392
0 .3898
2
362
52.3
405
89.8
0.6305
0 . 8245
12.9777
0.0635
2
363
52.3
405
90.8
0.49 84
0.8332
3.3250
0.2506
2
3 6 4
52.3
405
48.7
0.5691
0.4471
7.4243
0.0602
2
365
52.3
405
48.0
0.6244
0.4404
12.3431
0.0357
2
366
52.3
410
47.9
0.4309
0.4393
1 . 2784
0 .3436
2
367
52.3
405
137.0
0.5100
1.2571
3 . 8435
0 .3271
2
368
52 . 3
410
137.1
0.4309
1.2578
1 . 2784
0.9839
2
369
52.3
400
136.3
0.5691
1.2507
7.3129
0.1710
2
370
52 . 3
403
90.9
0.547 5
0.8339
5.8881
0.1416
2
371
52.3
426
137.1
0.5738
1.2582
B . 2811
0.1519
2
372
52.3
425
138.7
0 . 5438
1.2732
6.0652
0.2099
1
373
52.3
425
90.9
0.5325
0.8346
5.3562
0.1558
2
374
52 . 3
4 30
90.3
0 .4913
0.8288
3.3181
0.2498
1
37 5
52.3
400
91.4
0 .4618
0.8388
1.9765
0 .4244
1
BURN = 2 	 STABLE OPERATION
BURN = 1 	 STABILITY LIMIT
BURN = 0 	 BURNER GOES OUT
UNITS OF INT ARE . LBS/SEC FT*3 ATM*2
<
t-H ftj
n tr
•L *-
. ft

-------
COMPARISON OF PREDICTED AND EXPERIMENTAL BURNER STABILITY DATA	PAGE 16
RUN
NO.
VOL
IN* 3
AIR TEMP
OF
AIR FLOW
L3S/HR
EQUIV
RA TIO
INT DATA
W/VP*2
INT LIM
1/1 /VP* 2
INT RAT
ID/IL
BURN
376
52.3
400
90. 3
0.4711
0.8288
2.2577
0.3671
377
52.3
405
48.3
0.4838
0.4432
2.7404
0 .1617
378
52.3
4 2 8
45.5
0.4875
0.4173
3.1372
0.1330
379
52.3
375
47.0
0 . 4650
0 .4311
1.8754
0 . 2299
380
52.3
412
45.7
0.4395
0.4198
1.4789
0.2839
381
52.3
411
130.5
0 .4507
1 . 2710
1.7 498
0.7264
382
52.3
411
138.7
0.4501
1.2728
1.7336
0.7342
383
52.3
403
47.1
0.4725
0 .4320
2.3309
0.1853
381
52.3
76
91.4
0 . 5775
0.8390
2.8643
0.2929
385
52.3
90
91 . 6
0.5625
0.8410
2.4911
0.3376
386
52.3
91
89.8
0.5284
0.8245
1.5705
0.5244
387
52.3
80
4 7.8
0.5255
0.4389
1.4450
0.3034
388
52 . 3
81
45.5
0.5337
0.4180
1.6331
0.2557
389
"l .0
85
48. 7
1.2334
0.0000
0.0000
0 .0000
390
52.3
90
80.9
0 . 5400
0.7428
1.8470
0 .4021
391
52.3
93
80.5
0.4900
0.7384
0.8741
0 . B439
BURN = 2 	 STABLE OPERATION	UNITS OF INT ARE LBS/SEC FT*3 ATM*2
BURN = 1 --- STABILITY LIMIT
BURN =0 	 BURNER GOES OUT
i
h->
00
X1
Ol
er

-------
COMPARISON OF PREDICTED BURNER INEFFICIENCY AND EXPERIMENTAL EMISSIONS DATA
PAGE 1
RUN
BURN
AIR
FUEL
AIR

EQUIV
COMB

PRED
PRED
COG

RCG

NOBG
NOTG
NO.

TEMP
TEMP
FLOW

RA TIO
TEMP

EFFY
EMM
EMM

EMM

EMM
EMM


OF
OF
LBS/HR


OF


G/KG
G/KG

G/KG

G/KG
G/KG
1
"l
7 5
71
22.3
0
.4678
2107
0
.9671
32.8552
"l . 0000
"1
.0000
"1
.0000
"1 . 0000
2
"1
80
71
22.2
0
. 5309
2384
0
.9893
10.7091
"l . 0000
"1
.0000
~1
.0000
"l . 0000
3
~1
85
72
45.3
0
.4822
2098
0
.9325
67.5205
"l .0000
1
.0000
~1
. 0000
~1.0000
i*
1
85
73
4 5.3
0
.4 524
1924
0
. 8970
103.0000
"l . 0000
"1
. 0000
"l
.0000
"1 .0000
5
~1
85
73
47.0
0
. 4650
1954
0
. 8900
110 .0000
"l . 0000
~1
.0000
~1
. 0000
~1 .0000
6
"l
88
74
61 . 8
0
. 4501
1919
0
. 8970
103.0000
"l . 0000
"1
. 0000
~1
.0000
"1.0000
7
"1
90
75
61.7
0
. 49 48
2108
0
.9141
85.8886
"l .0000
"1
. 0000
~1
.0000
~1 . 0000
8
2
90
73
61.7
0
.6326
2742
0
.9898
10.1917
1 .0000
"1
. 0000
"l
.0000
"l .0000
9
1
90
74
61.7
0
. 5073
221 5
0
.9451
54.9107
"l.0000
~1
.0000
"1
.0000
"1 . 0000
10
0
90
74
61 . 7
0
. 4697
1974
0
. 8900
110.0000
"l .0000
"l
. 0000
"l
.0000
"l . 0000
1 1
1
90
74
61.7
0
. 5230
2303
0
.9611
38.9454
"l.0000
~1
.0000
"l
.0000
"1 .0000
12
0
90
74
61 . 7
0
. 43 22
1864
0
.9000
100 .0000
"l . 0000
"1
.0000
~1
.0000
"l .0000
13
~1
90
74
61 .7
0
. 5386
2378
0
.9701
29.8793
~1 .0000
"1
. 0000
"l
.0000
"l .0000
1 4
1
90
75
61.7
0
. 5073
2215
0
.9451
54.9107
"l .0000
"l
.0000
~1
.0000
"1 .0000
1 5
2
90
74
61 . 7
0
. 5089
2225
0
.9473
52.7360
"l .0000
"1
.0000
"l
.0000
"l .0000
16
2
90
75
78.1
0
. 5689
2491
0
.9743
25.7034
"l . 0000
"1
. 0000
~1
.0000
~1 .0000
17
1
90
75
78 .1
0
. 5417
2369
0
.961 5
38.5179
"l . 0000
"l
. 0000
"l
.0000
"l.0000
18
2
90
73
7 8.1
0
. 5 541
2427
0
.9662
31.7835
1.0000

.0000
~1
.0000
"l .0000
19
2
91
73
78.0
0
.6115
2659
0
.9841
15.9106
"l . 0000
~1
.0000
"l
. 0000
"1.0000
20
2
92
74
78.0
0
. 5377
2352
0
.9591
40 . 8980
"l . 0000
~1
.0000
"l
.0000
"l.0000
21
0
92
74
78.0
0
. 4609
1946
0
.8900
110.0000
"1.0000
"1
. 0000
"1
.0000
"1 .0000
22
0
93
72
77.9
0
.4960
2063
0
. 8890
111. 0000
~1.0000
"l
.0000
~1
.0000
"l.0000
23
1
93
72
77.9
0
. 5209
2258
0
.9421
57.9394
1.0000
~1
. 0000
1
.0000
"l.0000
24
1
93
74
77.9
0
. 53 82
2356
0
.9597
40 . 2943
"l .0000
"l
.0000
~1
.0000
"1 .0000
25
1
93
7 5
77.9
0
. 5631
2470
0
.9726
27 .4482
~1 . 0000
~1
.0000
~1
. 0000
~1.0000
26
2
90
72
103.3
0
. 6937
2934
0
. 9887
11 . 3444
"1 .0000
~1
. 0000
~1
.0000
"l.0000
27
2
93
73
103.1
0
. 5794
2517
0
.9688
31.1919
~1 .0000
~1
. 0000
~1
. 0000
"l .0000
28
0
96
73
102.8
0
. 5227
2133
0
.8800
120.0000
~1 .0000
~1
.0000
~1
.0000
"l .0000
29
0
97
73
102.7
0
.5118
2114
0
. 8870
113 .0000
~1 . 0000
"l
.0000
"1
. 0000
~1 .0000
30
1
97
74
102.7
0
. 5231
2136
0
. 8800
120.0000
~1 . 0000
~1
. 0000
~1
.0000
"l.0000
BURN = 2	STABLE
BURN-1	LIMIT
aURN = 0	GOING OUT
PRED EMM = 1000*(1-EFFY)
PRED EFFY FROM THEORY
NOB - BURNER DATA FOR NOX
NOT - NOX FROM VAPOR GENERATOR EXH,
>-«
M C7
I H-
!-•
10

-------
COMPARISON OF PREDICTED BURNER INEFFICIENCY AND EXPERIMENTAL EMISSIONS DATA
PACE 2
RUN
BURN
AIR
FUEL
AIR

EQUIV
COMB

PRED
PRED
COG
HCG
NOBC
NO.

TEMP
TEMP
FLOW

RA TIO
TEMP

EFFY
EMM
EMM
EMM
EMM


OF
OF
LBS/HR


OF


G/KG
G/KG
G/KG
G/KG
31
2
102
75
120.9
0
. 5930
2569
0
.9690
31.0396
"l.0000
"1.0000
"1.0000
32
2
105
75
120.6
0
. 5465
2339
0
. 9353
64.7047
~1 . 0000
~1.0000
"l.0000
33
1
108
77
120.2
0
.5158
2120
0
. 8800
120.0000
~1.0000
"l . 0000
"l.0000
34
0
95
75
121.6
0
. 5099
2106
0
. 8870
113.0000
"l .0000
1 . 0000
1.0000
3 5
1
102
73
120.9
0
.5190
2126
0
. 8800
120.0000
"l.0000
~1.0000
"l . 0000
36
2
112
74
153.8
0
. 5878
2530
0
. 9563
43 .7293
~1.0000
~1 . 0000
"l .0000
37
2
117
74
153.2
0
. 5613
2392
0
.9325
67 .4910
~1.0000
~1.0000
~1.0000
38
2
1 20
74
152.8
0
. 5451
2221
0
. 8790
121.0000
~1.0000
"l.0000
~1 . 0000
39
0
120
75
152.8
0
. 5299
2175
0
. 8800
120 .0000
~1.0000
"l.0000
"l.0000
40
~1
120
75
152.8
0
.5135
2122
0
. 8800
120 .0000
~1 .0000
"l . 0000
"l.0000
41
1
1 24
75
152.3
0
. 5470
2282
0
. 9021
97 .9472
~1.0000
"l . 0000
"1.0000
42
2
91
75
103.2
0
.7565
3127
0
.9915
8.5202
2.3336
1.0000
1.0000
43
1
93
75
103.1
0
. 5222
2140
0
. 8850
11 5 .0000
~1.0000
"l.0000
"l.0000
44
2
87
73
41.8
0
. 7529
3127
0
.9965
3 . 4990
"l .0000
~1.0000
~1.0000
45
2
85
75
41.9
0
.6916
2941
0
. 9954
4.5711
~1.0000
"1.0000
"1.0000
46
2
84
75
41.9
0
.6132
2678
0
.9918
8.1577
~1.0000
"l.0000
"l.0000
47
2
82
75
42.0
0
. 5333
2373
0
.9791
20.8626
~1.0000
"l.0000
"l.0000
48
2
82
74
42.1
0
.4910
2170
0
. 9542
4 5 . 8251
0.2371
1.0000
"1 . 0000
49
0
80
72
42.1
0
.4497
1913
0
. 8980
102 .0000
22.3454
"l . 0000
"l .0000
50
1
80
73
42.1
0
.4997
2215
0
.9619
38.1283
10.1863
"l . 0000
~1 . 0000
51
2
83
73
25.4
0
.6097
2673
0
. 9950
5 . 0074
~1.0000
~1 . 0000
"l.0000
52
2
82
73
25.4
0
. 5406
2420
0
. 9892
10 .7788
~1.0000
~1.0000
"l.0000
53
1
82
73
25.4
0
.4645
2083
0
.9584
41.5690
"l . 0000
"l . 0000
~1 . 0000
54
1
83
73
25.4
0
.4510
1992
0
.9356
64.3729
~1 . 0000
"l.0000
~i .oono
55
~1
83
73
42.0
3
.17 53
149?
0
.0000
0.0000
~1 . 0000
"l . 0000
~1 .0000
56
~1
81
73
42.1
2
.6412
1798
0
. 0000
0 .0000
~1.0000
"l .0000
"1 . 0000
57
~1
82
73
42.0
2
.8966
1639
0
. 0000
0.0000
6.4517
~1.0000
"l . 0000
58
"l
8 5
75
41.9
2
.6510
1794
0
.0000
0.0000
109 . 9697
"l . 0000
~1 .0000
59
~1
85
75
41.9
3
.1812
1490
0
.0000
0.0000
50 . 9444
"1 . 0000
1.0000
60
~1
90
75
41.7
1
. 9452
2447
0
. 0000
0.0000
53 . 2602
"l.0000
~1 . 0000
BURN = 2	STABLE	PRED EMM = 1000 *( 1-EFFY)	NOB - BURNER DATA FOR NOX
BURN=1	LIMIT	PRED EFFY FROM THEORY	NOT - NOX FROM VAPOR GENERATOR EXH.
BURN=0	GOING OUT
NOTG
EMM
G/KG
"1 . 0000
~1 .000A
"l.0000
"l.0000
~1.0000
~1 . 0000
1 .0000
"l . 0000
~1 . 0000
"l .0000
"l . 0000
"1.0000
"l.0000
"l.0000
"1.0000
"1.0000
~1 .0000
1 .0000
~1 . 0000
"l . 0000
"l . 0000
"l .0000
"l . 0000
"l.0000
"l.0000
"l .0000
"l . 0000
"l .0000
"l . 0000
"l . 0000
Table
VIII-20

-------
COMPARISON OF PREDICTED BURNER INEFFICIENCY AND EXPERIMENTAL EMISSIONS DATA
PACE 3
RUN
BURN
AIR
FUEL
AIR
EQUIV
COMB
PRED
PRED
COG
RCG
NOBG
NOTG
NO.

TEMP
TEMP
FLOW
RATIO
TEMP
EFFY
EMM
EMM
EMM
EMM
EMM


OF
OF
LBS/HR

OF

G/KG
G/KG
G/XG
G/KG
G/KG
61
~1
90
75
41 .7
1 .9452
2447
0.0000
0 .0000
12 54 . 97 20
"l . 0000
"l .0000
~1 . 0000
62
"1
85
72
41 .9
2.6510
1794
0.0000
0.0000
1135.7524
"l .0000
0.0013
"1.0000
63
"1
87
72
41.8
3 .1870
1 4 8 B
0.0000
0.0000
3.5209
"l .0000
0.0023
~1.0000
Sit

85
72
41.9
2.6510
17 9 4
0.0000
0.0000
910.4047
~1.0000
0.0004
~1.0000
65
~1
88
75
41.8
2.7692
1717
0.0000
0 .0000
"l . 0000
~1 . 0000
"l.0000
"l.0000
66
"l
8 5
75
41.9
2.8585
1661
0.0000
0 .0000
~1.0000
"l.0000
~1 .0000
"l.0000
67
~1
85
75
41.9
2 .7616
1720
0 .0000
0.0000
284.034 5
"l.0000
0.0013
~1.0000
68

78
7 5
25.5
2.5262
1875
0 . 0000
0 .0000
"l .0000
"l.0000
"1.0000
1.0000
69
"l
82
75
24.7
2.3985
1978
0 . 0000
0.0000
1272.6096
122.4420
0.0016
~1 . 0000
70
~1
85
78
25.3
2 .9167
1629
0 . 0000
0.0000
1041.4344
472.0485
0.0007
~1.0000
71
~1
8 6
78
25.3
3.0646
1549
0 .0000
0.0000
~1 .0000
"l . 0000
"l . 0000
~1.0000
72
"l
90
78
61 . 7
2 .3863
1993
0 .0000
0.0000
1200 .7858
151.4368
0.0006
~1 . 0000
73

93
78
61 . 5
2.4493
194?
0.0000
0.0000
~1.0000
~1 . 0000
"l.0000
"l.0000
74
1
86
73
25.3
0.5044
2278
0. 9821
17.9200
1.2104
9.5854
0.0014
~1.0000
75
2
85
73
25.3
0.6219
2717
0 . 9955
4.4678
0.1944
"l.0000
0.1023
~1.0000
76
2
85
7 5
25.3
0.8339
3339
0.9981
1.9035
0.3193
4.9987
0.4057
~1.0000
77
2
85
75
25.3
0 . 9960
3558
0.9877
12.3041
6.6866
2.3294
0.3974
"1,0000
78
1
90
75
61.7
0 . 5574
2459
0 .9774
22.6080
101.7 531
57.3660
0.0119
1.0000
79
2
92
79
61 .6
0 .6869
2926
0.9931
6.8677
0.9586
1.4905
0.2489
1.0000
80
2
90
79
61 .7
0.9927
3555
0.9812
18.7995
21.5030
"l.0000
0.3622
1.0000
81
~1
92
80
61.6
1.4432
3143
0 .0000
0 .0000
688.0142
"l.0000
0.0000
1.0000
82
2
103
75
102.1
0 . 5050
2096
0 . 8B70
113.0000
105.3359
7 .7417
0 .0000
~1.0000
83
2
105
78
101.9
0.7666
3165
0.9920
7.9883
2.7654
3.4637
0.3348
"1.0000
84
2
100
72
102.4
0.7464
3104
0.9914
8.6103
3.5505
3.9463
0.4516
~i.oooo
85
2
103
77
102.1
0.9178
349 3
0.9 899
10.1368
12.6460
1.3655
0.4084
~1 . 0000
86
~1
105
77
101.9
1.2812
3502
0 . 0000
0.0000
866.5566
0.9724
0.0264
~1.0000
87
2
300
600
59 . 3
0.5391
2569
0.9859
14.0724
0.13 44
1.5420
0 . 0411
"1.0000
88
2
300
600
S9 . 3
0.6869
3071
0.9955
4,4707
1.2088
0 .2238
0.3218
~1 . 0000
89

300
600
59 . 3
1.1075
3658
0 .0000
0 . 0000
476.8417
0.8592
0.0356
"l.0000
90
2
300
720
59 .6
0.5770
2708
0 .9904
9 . 5539
0.4080
0.5712
0.2221
"l . 0000
BURN*2	STABLE	PRED EMM = 1000x(1 -EFFY)	NOB - BURNER DATA FOR NOX
BURN=1	LIMIT	PRED EFFY FROM THEORY	NOT - NOX FROM VAPOR GENERATOR EXH.
BURN =0	GOING OUT
Table
VIII-21

-------
COMPARISON OF PREDICTED BURNER INEFFICIENCY AND EXPERIMENTAL EMISSIONS DATA
PAGE 4
RUN
BURN
/llfl
FUEL
,41/?
EQUIV
COMB

PRED
PRED
COG
HCG
NOBG
NOTG
NO.

TEMP
TEMP
FLOW
RATIO
TEMP

EFFY
EMM
EMM
EMM
EMM
EMM


OF
OF
LBS/HP

OF


G/KG
G/KG
G/KG
G/KG
G/KG
91
2
310
720
59 . 3
0.8222
3440
0
. 9967
3.3125
1 . 7928
0.0851
0.3795
"l . 0000
92
~1
300
720
59 . 3
1 .1075
, 3658
0
.0000
0 . 0000
699 .7134
0.1234
0.0915
"1.0000
93
2
250
4 40
165.3
0.4863
2176
0
.9011
98 .8889
51.3433
~1 .0000
0.0128
~1.0000
94
2
250
485
165.3
0 .6478
2900
0
.9890
11.0026
4.0006
~1.0000
0.1154
"l . 0000
95
2
250
510
165.3
0 . 8299
3416
0
. 9937
6.2930
9.3111
"l .0000
0.5037
"l.0000
96
2
250
520
132.9
0 .9865
363 6
0
.9817
18.2715
207 . 8635
~1 .0000
0.4010
"l.0000
97
2
95
77
55.9
0.48B1
2192
0
.9648
35.1657
"1.0000
"1.0000
0 . 3787
1.0000
98
2
9 5
77
55.9
0.6175
2705
0
. 9936
6 . 3709
"l.0000
"l .0000
0 .4642
"l.0000
99
2
9 5
77
55.9
0.8087
3283
0
.9974
2.6068
~1.0000
"l.0000
0.4747
~1.0000
100
2
100
77
55.6
0.9984
3565
0
.9841
15.8934
~1.0000
~1.0000
0.3623
~1.0000
101
~1
102
77
55.5
1.1925
3485
0
.0000
0.0000
"l .0000
"l .0000
0.3553
"1.0000
102
2
102
77
89.0
0.7520
3132
0
. 9954
4 . 5716
~1.0000
1.0000
0 . 4443
"1.0000
103
2
400
770
50 . 0
0.8359
3525
0
.9984
1.6190
~1 .0000
"l.0000
~1.0000
~1 . 0000
104
2
400
800
50 .0
0.5406
2666
0
.9946
5.4009
"l .0000
"1.0000
"1 .0000
"1 . 0000
105
2
84
88
49.9
0.9250
3 503
0
.9964
3.6101
~1 . 0000
"l .0000
~1.0000
~1 .0000
106
1
8 5
95
49.9
0.4888
2194
0
.9687
31 . 2602
3.9308
~1 .0000
0.0049
"l . 0000
107
0
85
89
49.9
0.3925
1729
0
. 9060
94 .0000
"l .0000
"l.0000
"l.0000
~1.0000
108
2
8 5
92
49 .9
0.4448
1900
0
. 8980
1 02 . 0000
~1.0000
"l.0000
~1.0000
~1.0000
109
2
85
9 8
49.9
0.5563
2477
0
.9890
11.0324
~1.0000
"1.0000
1.0000
~1.0000
110
2
400
730
50 . 0
0 .4039
2115
0
.9606
3 9.3993
16.3425
"l.0000
0.0030
~1.0000
111
2
400
730
50 .0
0.3915
20t4
0
.9450
55.0286
47.2658
"l.0000
0.0832
~1.0000
112
2
400
720
33.2
0.4688
2407
0
.9919
8 .0925
0.1555
"l . 0000
0.7380
"1.0000
113
2
400
725
33.2
0.3891
2072
0
. 9677
3 2 .3190
17.5557
"l.0000
0.1116
~1.0000
114
2
400
730
33.2
0.380 5
2077
0
. 9598
40.1876
11.5914
"l.0000
0.2476
"1.0000
115
1
400
720
33.1
0.3883
2068
0
. 9672
3 2.8187
~1 .0000
"l.0000
0.0186
~1 .0000
116
2
400
710
82.7
0.537 5
2647
0
. 9906
9.3964
"l.0000
"l.0000
~1 .0000
~1 .0000
117
2
400
710
82.8
0.4312
2209
0
. 9573
4 2.7243
~1.0000
~1 . 0000
~1 . 0000
~1.0000
118
0
400
730
82.8
0.3 861
1916
0
. 8790
121.0000
68.5008
"l.0000
0.0125
"1 . 0000
119
2
400
720
126.8
0.4672
2346
0
.9623
37.7045
1.0000
1.0000
0.0308
~1.0000
120
2
400
475
127.0
0.4279
2041
0
. 8700
130.0000
~1.0000
"l .0000
~1.0000
~1.0000
BURN - 2	STABLE
BURN = 1	LI11IT
'RN = 0	GOING OUT
PRED EI'M = 1000x(i-EFFY)
PRED EFFY FROM THEORY
NOB - BURNER DATA FOR NOX
NOT - 1
-------
COMPARISON OF PREDICTED BURNER INEFFICIENCY AND EXPERIMENTAL EMISSIONS DATA
PACE 5
RUN BURN
AIR
FUEL
AIR
EOtllV
COr'B

PRED
PRED
COC
HCG
NOBG
NOTG
NO.

TEMP
TEMP
FLOW
RATIO
TEl'P

EFFY
E!'M
Ei-ii-;
EMM
EMU
EMM


OF
OF
LBS/HR

OF


C/KC
C/KC
G/KG
G/KG
G/KG
121
2
400
4 50
126.8
0.4469
2239
0
. 9432
56.7651
~1 . 0000
~1 . 0000
~1.0000
~1 . 0000
122
0
400
4 50
126.8
0.4010
1963
0
. 8770
123.0000
"l . 0000
~1 . 0000
~1 . 0000
"l . 0000
123
2
74
70
50 . 5
0.7951
3236
0
. 997 5
2.4 595
0.3215
~1 . 0000
0.4284
"l . 0000
124
1
80
78
49.5
0 .4743
2112
0
. 9 5 56
44.4243
13.8280
~1 . 0000
0.0757
1.0000
125
1
8 5
78
47,6
0.4930
2218
0
. 9728
27 . 2300
17.2647
"l.0000
0.0582
"l . 0000
126
0
9 5
7 8
127.3
0.5049
2092
0
. 8880
112 . 0000
12.3792
~1.0000
0.2459
"l.0000
127
1
9 5
78
119.6
0.5687
2498
0
. 97 57
24.3086
16.0140
~1.0000
0.2088
1.0000
128
2
90
100
67.8
0.5411
2411
0
.9815
18.4842
~1.0000
~1.0000
~1.0000
"l.0000
1 29
2
80
80
86.2
0.5061
2217
0
.9515
4 8.4529
"l.0000
~1.0000
~1.0000
"l .0000
130
2
80
80
82.8
0.5822
2556
0
.98 56
14.3664
"1.0000
~1.0000
~1 .0000
~1 . 0000
131
2
80
80
66 . 7
0.6869
2923
0
. 99 52
4.7632
"l.0000
~1.0000
~i.onoo
"l . 0000
132
~1
80
80
4 4.5
1.0553
3562
0
. 0000
0.0000
"l.0000
~1.0000
~1.0000
~1 . 0000
133
2
80
80
83.4
0.5628
2480
0
.9819
18.0994
"l . 0000
"1.0000
"1.0000
1.0000
134
2
80
80
66.7
0 .7036
297 5
0
. 99 57
4.3442
~1 . 0000
~1 . 0000
~1 . 0000
~1 . 0000
135
0
80
80
72.3
0.4347
1864
0
.9000
1 00 . 0000
"l . 0000
38.1951
~1.0000
"l . 0000
1 36
2
88
90
72.3
0.5318
2367
0
.9770
23.017 5
~1 . 0000
95.3023
1.0000
"l .0000
137
2
90
90
66 . 7
0 .7036
2982
0
.99 57
4.2635
"l .0000
0.9204
~1.0000
~1 . 0000
138
2
9 5
95
96.6
0.6404
2773
0
. 9907
9.3221
"l . 0000
~1 . 0000
0 . 3612
"l .0000
139
2
9 5
1 00
95.0
0 .7260
3 0 50
0
. 9946
5.4436
"l .0000
"1.0000
~1 . 0000
1.0000
140
~1
9 5
100
91.1
1.0266
"3 569
0
.0000
0.0000
"l .0000
~1 . 0000
~1 . 0000
"l .0000
141
~1
95
100
77.3
1.2099
3460
0
.0000
0.0000
1 .0000
~1.0000
~1.0000
"l.0000
142
2
9 5
100
77.8
0.5389
2399
0
.97 80
22.0132
"l . 0000
~1.0000
~1 . 0000
"1.0000
143
~1
9 5
100
77.8
1.2013
347?
0
.0000
0.0000
"l.0000
~1.0000
~1.0000
"l . 0000
144
"l
95
100
91.1
1.0266
3569
0
. 0000
0.0000
1.0000
"l.0000
~1.0000
"l.0000
145
2
9 5
1 00
94.4
0.7302
3062
0
. 9947
5.3162
"l.0000
"l.0 0 00
1 .0000
1.0000
146
2
95
100
96 . 6
0.5433
2404
0
.9732
26.8288
"l.0000
~1.0000
~1.0000
"l . 0000
147
1
95
100
98.3
0.470?
1979
0
. 8900
110.0000
~1 . 0000
~1.0000
~1.0000
"l .0000
148
2
9 5
100
90 . 5
0.7614
3153
0
. 9954
4.6212
"l . 0000
~1 . 0000
"1.0000
1.0000
149
~1
9 5
1 00
53.6
1.2873
3 356
0
.0000
0.0000
"l . 0000
~1.0000
0.0000
~1 . 0000
1 50
2
95
1 00
56.9
0.4388
1887
0
. 3980
102.0000
1.0000
"l.0000
~1 . 0000
~1 .0000
BURN-2
_ _ _
STABLE

PRED EMM
= 1 00 0 x ( 1
-EFFY)

NOB -
BURNER DATA FOR NOX


BURN=1
	
L Ii'il T

PRED EFFY FROtt THEORY


NOT -
NOX FROM
VAPOR GENERATOR EXH .

3URN=0
	
GOING
OUT










Table
VIII-23

-------
COMPARISON OF PREDICTED BURNER INEFFICIENCY AND EXPERIMENTAL EMISSIONS DATA
PAGE 6
RUN BURN
AIR
FUEL
AIR

EOUIV
COMB

PRED
PRED
COG
HCG
NOBG
NOTG
NO .

TEMP
TEMP
FLOV
RA TIO
TEfi'P

EFFY
EMM
EMM
EMM
EMM
EMM


OF
OF
LBS/HR


OF


G/KG
G/KG
G/KG
G/KG
G/KG
1 51
2
400
160
84.0
0
. 7066
3197
0
.9969
3.1313
~1.0000
"l.0000
~1 .0000
"l.0000
152
2
400
330
79.5
0
.7461
3308
0
. 9974
2.6398
~1 . 0000
0.1098
~1.0000
"l.0000
1 53
1
360
300
111.2
0
.4703
2329
0
.9642
35.8068
1.0000
0 .0000
~1 . 0000
~1.0000
1 54
2
355
200
55.6
0
.7013
3155
0
.9977
2.2649
~1.0000
"l.0000
"l.0000
~1.0000
155
2
270
275
98.9
0
. 5043
2401
0
.9743
25.7017
~1 .0000
0.1179
~1 . 0000
"l.0000
156
0
260
255
103.3
0
.4292
1963
0
. 8860
114.0000
~1 . 0000
"l.0000
~1 . 0000
~1.0000
157
1
260
290
100.0
0
. 5856
2706
0
. 9896
10.3856
~1 .0000
0.0099
"l.0000
"l . 0000
158
2
260
317
100 .6
0
.637 5
2884
0
.9931
6.8859
~1 . 0000
0.0700
~1.0000
"1.0000
159
2
265
3 50
97 . 3
0
. 7598
3257
0
.9962
3 .8282
"l . 0000
0.0621
~1.0000
~1.0000
160
~1
270
410
90 .7
1
. 2226
3546
0
.0000
0.0000
~1 . 0000
"l.0000
~1.0000
"l.0000
161
~1
275
39 2
89 . 6
1
. 2418
3525
0
. 0000
0.0000
~1 . 0000
0.0789
~1.0000
~1.0000
162
"l
275
480
87.9
1
. 5592
3096
0
. 0000
0 .0000
~1 . 0000
6.4648
"1.0000
"l.0000
163
1
360
110
22.0
0
. 5464
2663
0
.997 6
2.4453
~1.0000
63.1902
1.0000
1.0000
164
2
370
120
22.0
0
.7 565
3321
0
.9993
0.7427
~1.0000
28.6829
"l . 0000
"l.0000
165
2
375
120
30.8
0
.7925
3415
0
.9990
0.9942
"l.0000
0.0721
~1.0000
"l.0000
166
2
365
140
44.5
0
.6641
3051
0
. 9979
2.1090
1 . 0000
0 .0337
~1.0000
"l .0000
167
2
370
150
44.0
0
. 8406
3518
0
.9985
1.4867
"l.0000
0.0336
"l.0000
"1 .0000
168
2
370
170
41.2
0
.9728
3701
0
. 9950
4.97 34
"1 . 0000
0.0649
"1 .0000
"1.0000
169
2
345
1 80
55.0
0
.7296
3230
0
.9979
2.0617
~1.0000
0.0260
~1 . 0000
"l.0000
170
2
325
190
69.8
0
. 5771
2733
0
.993 5
6 . 5443
~i.oooo
0.0051
~1 . 0000
"l.0000
171
2
305
195
78.1
0
.5162
2491
0
.98 51
14.8941
1.0000
0.0000
"1.0000
1.0000
172
0
275
195
97.8
0
.4118
1917
0
. 8870
113.0000
"l . 0000
1.0630
"l . 0000
"1.0000
173
~1
"l
"l
"1.0
1
.0000
4.
0
. 0000
0.0000
"l . 0000
"l.0000
~1.0000
"l.0000
174
2
400
250
96 .6
0
.6507
3027
0
.9954
4.6110
~1 . 0000
0.0261
"1 . 0000
~1.0000
17 5
2
405
275
93 .3
0
.7925
3428
0
.9971
2.9001
~1 . 0000
0 . 0291
"l.0000
"l.0000
176
2
405
280
93.1
0
. 8733
3597
0
.9967
3.3 376
~1 . 0000
0.0308
"1 . 0000
"l.0000
177
2
405
200
98.6
0
. 4239
2145
0
. 9338
66.1992
~1 . 0000
0.0000
0.2043
"l.0000
17 8
2
405
195
9 5.9
0
. 5413
2661
0
.9896
10.4432
"l . 0000
0.0000
"l . 0000
~1.0000
179
2
350
300
133.7
0
. 8388
3497
0
.9954
4.5738
1 . 0000
0.0088
1.0000
"l.0000
180
2
345
310
134.6
0
.7140
3178
0
.9947
5.3268
~1 . 0000
0.0151
~1 . 0000
"l.0000
BURN=2
	
STABLE

PR ED EMM
_
1000
*(1-EFFY)

NOB -
BURNER DATA FOR NOX


BURN*1
	
LIMIT

PRED EFFY
FROM
THEORY


NOT -
NOX FROM
VAPOR GENERATOR EXH.

BURN=0	GOING OUT
Table
VIII-2K

-------
COMPARISON OF PREDICTED BURNER INEFFICIENCY AND EXPERIMENTAL EMISSIONS DATA
PAGE 7
RUN
BURN
-4
FUEL
AIR
EQUIV
COMB

PRED
PRED
COG
HCG
NOBG
NO.

TEMP
TEMP
FLOW
RA TIO
TEMP

EFFY
EMM
EMM
EMM
EMM


OF
OF
LBS/ NR

OF


G/KG
G/KG
G/KG
G/KG
181
2
340
305
136.1
0.620 5
2880
0
.9911
8.9491
"l . 0000
0.0137
"1 . 0000
182
2
340
303
138.8
0.5458
2613
0
. 9821
17.8870
"l.0000
0.0108
~1 . 0000
183
2
33 5
295
139.0
0 .4894
2372
0
.9616
38.3977
"l.0000
0 .0062
~1.0000
184
1
334
280
141.7
0.4435
2053
0
. 8770
123.0000
1 . 0000
0 . 0069
0 .1083
185
2
335
320
139.0
0.5014
2427
0
. 9681
31.8877
"l.0000
0 .0240
~1.0000
186
2
335
340
13 9.0
0.563 1
2674
0
.9849
15.1181
"l.0000
0 .0000
0.4135
187
"l
"l
~1
"l .0
"l .0000
~1
0
.0000
0 . 0000
1.0000
"l.0000
"1 . 0000
188
2
430
340
77.9
0.6830
3149
0
. 9970
3.0325
"l.0000
0 .0077
"l . 0000
189
2
430
320
79 . 0
0 . 5532
2727
0
. 9929
7.1253
"l.0000
0.0000
"l . 0000
190
1
430
300
79 . 0
0 .4885
2488
0
.9856
14.3782
1.0000
0.0000
1.0000
191
2
400
295
66.9
0.5766
2788
0
.9947
5.2707
"l.0000
0.0148
"1 . 0000
192
2
405
295
56 . 0
0.6896
3153
0
. 9978
2.2005
"l . 0000
0.0039
"1 . 0000
193
"l
410
360
51 . 6
1 . 2 578
3587
0
.0000
0.0000
"l . 0000
23.7001
"1 . 0000
194
"l
410
420
113.0
0.5453
2679
0
.9883
11.6743
"l .0000
0.0314
"1.0000
195
"1
410
410
113.6
0 .485B
2446
0
. 9762
23.7640
~1 . 0000
0.0353
~1.0000
196
"l
410
390
115.2
0.49 20
2470
0
.9778
22.1823
"l.0000
8.9366
"l . 0000
197
"l
430
340
101.4
0.5304
2640
0
. 9884
11.5726
1.0000
0 .9200
"1 .0000
198
"l
350
260
41.2
0.6579
3028
0
. 9980
2.0329
~1.0000
0.0084
"l . 0000
1 99
"l
340
320
38 . 5
0.9818
3690
0
.9938
6.1932
"l.0000
0.0380
"l.0000
200
~1
410
270
38.5
0.6616
3074
0
.99 83
1.6980
~1.0000
0.5087
~1.0000
201
2
85
82
83.2
0.8221
3307
0
. 9961
3.9168
"l.0000
~1.0000
"l.0000
202
2
8 2
82
102.9
0 .7187
3017
0
. 9938
6.2446
"l.0000
"l.0000
"l.0000
203
2
110
100
148.0
0.6216
2704
0
.9032
16.7514
1.8195
0.0278
0.3728
204
"l
~1
~1
"1 . 0
"l .0000
"l
0
. 0000
0.0000
"l.0000
"l.0000
"l.0000
205
2
313
77
76 . 8
0 .6978
3114
0
.9966
3.3925
"l.0000
"l.0000
~1.0000
206
2
253
77
167 . 8
0.6498
2909
0
.9890
1 0 . 9588
1.0000
1.0000
"1 .0000
207
2
245
77
181.5
0.5601
2572
0
.9728
27.2419
"l.0000
"l .0000
0.2588
208
1
70
~1
51 . 5
0.8177
3292
0
.99 80
1.9506
~1 . 0000
"l . 0000
"l .0000
209
"l
70
~1
49 .9
0.98B3
3549
0
.9904
9.5711
~1.0000
"l . 0000
~1 . 0000
210
~1
70
~1
49 . 9
0.6666
2856
0
.9968
3.2151
"l.0000
"l . 0000
"l . 0000
BURN =2	STA3LE	PRED EMM = 1 000*( 1-EFFY)	NOB - BURNER DATA FOR NOX
BURN = 1	LIMIT	PPED EFFY FROM THEORY	NOT - NOX FROM VAPOR GENERATOR EXH.
BURN = 0	GOING OUT
NOTG
EMM
G/KG
1 .0000
1 . 0000
1.0000
1.0000
1 . 0000
1.0000
1.0000
1.0000
1.0000
1.0000
1.0000
1.0000
1.0000
1.0000
1 .0000
1.0000
1.0000
1.0000
1 .0000
1.0000
1.0000
1.0000
1.0000
1 .0000
1.0000
1.0000
0.1527
1.0000
1.0000
'l .0000
Table
VIII-25

-------
COMPARISON OF PREDICTED BURNER INEFFICIENCY AND EXPERIMENTAL EMISSIONS DATA
PAGE 8
RUN
BURN
AIR
FUEL
AIR
EQUIV
COMB
PR ED
PR ED
COG
HCG
NOBG

NOTG
NO.

TEMP
TEMP
FLOW
RATIO
TEMP
EFFY
EMM
EMM
EMM
EMM

EMM


OF
OF
LBS/HR

OF

G/KG
G/KG
G/KG
G/KG

G/KG
211
"l
70
"l
49.9
1.0852
3 544
0.0000
0 . 0000
"l . 0000
~1 .0000
"l .0000
~1
. 0000
212
"l
70
~1
102.9
0.6368
2746
0.9915
8.4514
"l . 0000
~1.0000
~1 .0000
"l
. 0000
213
"1
70
~1
99.7
0 . 7865
3208
0.9961
3.9081
1 .0000
"l . 0000
"l . 0000
~1
. 0000
214
"l
70
"l
96.6
1.0999
3 537
0.0000
0.0000
"l . 0000
"l . 0000
"l . 0000
~1
. 0000
215
"l
7 0
~1
49.9
0.4845
2175
0.9729
27.0681
~1.0000
~1.0000
1 . 0000
1
. 0000
216
"l
70
~1
4 8.2
0.6615
2839
0.9968
3.2156
"l .0000
"l .0000
"l . 0000
~1
. 0000
217
2
80
~1
83.8
0,6044
2638
0.9885
11.4746
~1.0000
~1 .0000
"1.0000

. 0000
218
2
100
70
91.5
0.6187
2708
0.9919
8 .1171
0.1629
~1 .0000
0.0726
~1
. 0000
219
2
110
70
95.3
0.7845
3228
0.9965
3.5301
5.5606
~1 .0000
0.9701
~1
. 0000
220
2
100
100
86.2
0.5854
2583
0 . 9863
13.6626
1.0000
0.3391
0.2876
0
. 2431
221
2
100
100
116.8
0.5761
2534
0 . 9788
21.1727
"l.0000
0.3523
"l .0000
~1
. 0000
222
2
1 00
100
131.8
0.5105
2156
0.9078
92.1901
~1.0000
2.0954
1.0000
1
. 0000
223
2
100
100
108.4
0.6205
2704
0 . 9877
12.3488
~1.0000
0.5651
"1.0000
"l
.0000
224
2
100
100
168.5
0.5562
2413
0.9 555
44.5324
"l.0000
0.3447
~1 .0000
"l
. 0000
225
2
1 00
100
151.8
0.6173
2679
0.9817
18.2645
"l.0000
0.9205
"l .0000
"l
. 0000
226
1
100
1 00
177.9
0.5266
2149
0.8800
120.0000
~1.0000
"1.0000
"l.0000
~1
.0000
227
2
76
82
55.8
0.5796
25 54
0.9903
9.7 308
0.1744
0 .9677
0.7041
~1
. 0000
228
2
82
90
58.4
0.5761
254 5
0.9896
10.4349
0.1755
0.4170
0.0989
~1
. 0000
229
2
85
95
57.6
0.5773
2 552
0 .9899
10. 0640
0.1751
0 . 6529
0.3103
1
. 0000
230
2
433
98
57 . 6
0.5779
2818
0.99 59
4.1452
"l . 0000
0.4560
0.2053
"l
. 0000
231
2
420
1 00
60 . 8
0.5473
2702
0.9941
5.8817
~1.0000
0.3310
0.1174
~1
. 0000
232
2
410
100
90 .9
0.5203
2 589
0.9878
12.1522
1 . 0000
0.2877
0.1879
~1
. 0000
233
2
407
1 00
90.2
0.6172
2925
0.9948
5.2052
"l.0000
0.4614
0.5288
"l
. 0000
234
2
90
82
7E .9
0.6725
28 81
0.9941
5.9028
"l.0000
1.5728
1 .2813
1
. 0000
235
2
92
89
77.3
0.5930
2608
0.9 886
11.4415
"l .0000
2.6491
0.8953
"l
.0000
236
2
89
89
7 7.7
0.6003
2632
0 . 989 2
10.7661
~1 . 0000
2.9759
0.5918
"l
. 0000
237
2
82
82
38.4
0.63 53
2 7 C 0
0.9961
3.8535
1 .0000
2.9425
1.5835
1
. 9999
238
2
79
70
49.1
0.4880
2202
0.97 63
23.709 1
~1 . 0000
~1 .0000
0 .0637
"l
. 0000
239
2
80
71
48.9
0.5391
2414
0.9 894
10 . 5573
"l . 0000
1.0000
0.1855
"l
.0000
240
2
82
71
4 8.9
0.6563
2830
0.9967
3.2991
"l.0000
"l.0000
0.4526
"l
. 0000
BURN
= 2	
STABLE

PRED EMM
= 1000x(l
-EFFY)
NOB
BURNER DATA FOR NOX



BURN
= 1	
LIMIT

PRED EFFY FROM THEORY

NOT
NOX FROM
VAPOR GENERATOR EXH.


3URN = 0	GOING OUT
Table
VIII-26

-------
COMPARISON OF PREDICTED BURNER INEFFICIENCY AND EXPERIMENTAL EMISSIONS DATA
PACE 9
RUN
BURN
AIR
FUEL
AIR
EQUIV
COMB
PRED
PR ED
COG
HCG
NOBC

NOTG
NO.

TEMP
TEMP
FLOW
RATIO
TEMP
EFFY
Etlll
EMM
EMM
EMM

EMM


OF
OF
LBS/HR

OF

G/KG
G/KG
G/KG
G/KG

G/KG
241
2
82
72
48.8
0.7266
3051
0 .9978
2.2394
"l . 0000
~1.0000
1.9359
~1
.0000
242
2
78
68
49 . 2
0.7617
3150
0.9980
2.0079
~1.0000
~1.0000
1 .8018
"l
. 0000
243
2
80
72
48.8
0 . 8945
3457
0.9978
2.2083
~1.0000
~1.0000
2.6989
~1
.0000
24 4
2
79
68
98.1
0.5043
2221
0.9570
43 .0363
"l.0000
*1 .0000
0.1184
~1
. 0000
245
2
80
70
97 . 6
0.5994
2622
0 .9890
10.9873
~1.0000
~1 . 0000
0.2608
"l
.0000
246
"1
80
70
60 . 5
0.G393
2770
0.9953
4.6675
~1.0000
1.0000
1 . 0000
~1
.0000
247
"l
7 5
70
"l .0
0.4609
2131
0.0000
0.0000
"1.0000
"l.0000
~1 .0000
"l
.0000
248
~1
85
70
47.2
0.5625
2507
0.9925
7.4574
~1.0000
~1.0000
~1 .0000
~1
.0000
249
"1
7 5
70
49.2
0.8476
3363
0.9981
1.8957
~1.0000
~1 . 0000
0.1125
"l
. 0000
250
"l
85
70
48.8
0.8555
3386
0.9981
1.8842
"l.0000
"l.0000
1.2634
~1
. 0000
251
~1
85
70
100 . 8
0.4983
2190
0 .9502
49.7573
~1.0000
~1 .0000
0 .0767
"l
. 0000
252
"l
90
70
100 . 4
0.5006
2210
0.9543
45.7280
~1.0000
"l.0000
0.0501
"l
. 0000
253
"l
92
70
97.0
0.5772
2549
0.9 866
13.4016
~1 .0000
~1 . 0000
0.1275
"1
. 0000
254
"l
93
70
100 . 1
0.6412
2779
0.9925
7.5138
~1.0000
~1 .0000
0.2430
~1
.0000
255
2
93
70
90 .0
0.6063
2653
0.9882
11.7618
~1.0000
~1 . 0000
~1.0000
"l
.0000
256
2
85
70
97.6
0.7188
3024
0.9954
4 . 6069
~1.0000
~1.0000
0.4797
~1
. 0000
257
2
85
70
97 . 6
0.7852
3214
0.9963
3 .7465
~1.0000
~1.0000
1.3082
"l
.0000
258
2
85
70
97.5
0.9004
3466
0.9955
4 .4843
~1.0000
~1.0000
2.0235
~1
.0000
259
2
85
70
97 . 4
1.0000
3556
0 .9804
19.6038
~1.0000
~1.0000
2.6256
~1
.0000
260
2
85
70
97 . 3
0 .9609
3 538
0.9915
8.5162
~1.0000
"l.0000
0.9896
~1
. 0000
261
"l
89
70
97.2
1 .1875
3482
0.0000
0.0000
"l.0000
"l.0000
1.1944
"l
.0000
262
2
90
70
1 44 . 4
0.5167
2244
0.9436
56.4055
~1.0000
1.0000
0.2285
1
. 0000
263
2
97
70
143.5
0.5957
2609
0.983 5
16.4723
~1.0000
"l.0000
0 . 3739
~1
.0000
264
2
85
73
48.8
0.4951
2240
0.9799
20.0719
0.1469
"l . 0000
0.1062
~1
. 0000
265
2
85
73
48.8
0.5820
2577
0 . 9938
6.2279
0.1737
~1.0000
0.4157
"l
. 0000
266
2
96
70
143 .0
0 . 5677
2498
0 . 9772
22.8177
0.2547
"l.0000
~1 .0000
"l
. 0000
267
2
96
70
97 . 8
0.5045
2243
0.9611
38.9211
~1 . 0000
~1.0000
0 .0970
~1
.0000
268
2
96
70
97.2
0.5817
2 569
0.9874
12.6322
~1.0000
1.0000
0 . 2692
"1
.0000
269
2
96
72
97 . 2
0.6582
2838
0.9936
6.3814
"l.0000
"l.0000
0.6805
"l
. 0000
270
2
96
72
97 . 1
0.6860
2929
0.9947
5.3297
"l.0000
~1.0000
1.9372
"l
. 0000
BURN
= 2	
STABLE

PRED EMM
= lOOOxd
-EFFY)
NOB
BURNER DATA FOR NOX



BURN
= 1	
LI HIT

PRED EFFY FROM THEORY

NOT -
NOX FROM
VAPOR GENERATOR EXH.


BURN=0	GOING OUT
Table
VIII-27

-------
COMPARISON OF PPEDICTF.D BURNER INEFFICIENCY AND EXPERIMENTAL EMISSIONS DATA
PACE 10
RUN
BURN
j4 Jfl
FUEL


EQUIV
COMB

PPED
PRED
COG
HCG
NOBG
NOTG
NO.

TEMP
TEMP
FLOW

RATIO
TEMP

EFFY
F.'IM
EMM
EMM
EMM
EMM


OF
OF
LBS/MR


OF


G/KG
G/KG
G/KG
G/KG
G/KG
271
2
100
75
144.1
0
. 5576
24 59
0
.97 40
26.0316
~1 . 0000
~1.0000
0.1791
~1.0000
272
2
100
71
141.7
0
.6618
2846
0
.9909
9.1203
"l.0000
"1.0000
0.5020
"1.0000
273
2
100
70
143.1
0
. 6745
2888
0
.9915
8 .4565
"l.0000
~1.0000
1.4659
"1.0000
2 7 it
2
100
70
143.5
0
. 7 56 5
3140
0
. 9942
5.7548
0.4611
0.0000
0.386 5
~1.0000
275
2
95
70
143.7
0
. 8254
3317
0
. 9947
5.2594
0.7331
1.0000
1.0296
"1 . 0000
276
~1
~1
~1
130.0
"l
. 0000
"l
0
. 0000
0.0000
"l.0000
~1.0000
~1 . 0000
~1 . 0000
277
"l
~1
~1
140 .0
"1
.0000
1
0
. 0000
0 . 0000
"l . 0000
~1 . 0000
"l .0000
"1.0000
278
"l
~1
"l
"1 . 0
"l
. 0000
~1
0
. 0000
0.0000
"l .0000
"l . 0000
"l.0000
1.0000
279
2
109
70
91 .0
0
. 5791
2564
0
. 9848
15.1737
0.1247
0.1994
0.3360
0.1639
260
2
110
70
90 . 9
0
. 6808
2920
0
. 9936
6.3889
0.2104
0.9399
0 .7810
0.7188
281
2
110
70
111.3
0
. 5932
2610
0
. 9838
1 6 . 2248
0 .1216
0,0485
0.3316
0 .1638
282
2
94
70
46. 4
0
. 607 2
2670
0
. 9942
5.7850
0.1187
0 .0000
0 . 366 5
0 . 3002
283
2
100
70
47.3
0
. 60 24
2658
0
. 9939
6.0739
0 .1196
0.0000
0.1101
0.0786
2 8 4
2
105
70
46.5
0
. 4239
1846
0
. 9000
100.0000
0.1727
0 .0000
0 .0624
0.0454
285
2
116
70
136.5
0
. 4873
2051
0
. 8890
111.0000
0 . 2450
0.0000
~1 .0000
0.0 589
286
2
82
75
58 . 8
0
. 5010
2253
0
. 9771
22.8988
"l .0000
~1 . 0000
0 . 1097
"1.0000
287
2
92
70
144.0
0
. 5260
2302
0
. 9 54 8
45.1854
2.6202
0.0000
0 .1586
~1 . 0000
288
2
90
69
1 44 . 3
0
. 5654
2483
0
. 97 57
24.2540
1 .0000
0.0000
0.1849
"l . 0000
289
2
9 5
69
143.6
0
.6016
2629
0
. 9843
15.7393
~1 .0000
0.0000
0.2559
1 .0000
290
2
99
60
142.9
0
.7193
3029
0
. 9934
6.6129
0.3138
0.3568
0.1304
"1 . 0000
291
"l
73
68
66.0
1
.3984
3192
0
. 0000
0.0000
"l . 0000
~1 . 0000
0.6409
"l .0000
292
"l
7 5
70
49 .1
1
. 5625
2965
0
.0000
0 . 0000
0.1314
161 .4518
0.4629
~1 . 0000
293
1
82
70
47.1
1
.3 594
3 251
0
. 0000
0 .0000
1109.9767
1.0542
"1 . 0000
~1 . 0000
294
"l
90
70
51 .1
1
. 2578
3393
0
. 0000
0.0000
764.7702
2.7660
"1.0000
"1 . 0000
295
"l
72
70
49.2
1
.3203
3297
0
. 0000
0.0000
961 .8264
0.9149
1.1960
"l .0000
296
~1
86
73
48.6
1
.343 8
3274
0
.0000
0.0000
944 .061 5
0.6170
1.6858
"l .0000
297
"l
92
71
48.3
1
.4141
3183
0
.0000
0.0000
1152 . 1380
0.5784
1.0119
1.0000
298
"l
95
71
48.2
1
.2813
3364
0
.0000
0 . 0000
879 .4185
0.8733
2.1113
"l.0000
299
"l
96
71
87.2
1
. 3750
3238
0
. 0000
0 . 0000
934.6333
0.91S9
"1 .0000
"l.0000
300
"l
85
69
49.3
1
.1719
3494
0
.0000
0 .0000
1013.6513
1 .0314
0.8173
"l . 0000
BURN = 2	STABLE	PRED EMM = 1000 *( 1 -EFFY)	NOB - BURNER DATA FOR NOX
BURN=1	LIMIT	PRED EFFY FROM THEORY	NOT - NOX FROM VAPOR GENERATOR EXH.
BURN=0	GOING OUT
Table
VI11-28

-------
COMPARISON OF PREDICTED BURNER INEFFICIENCY AND EXPERIMENTAL EMISSIONS DATA
PACE 11
RUN
BURN
AIR
FUEL
AIR

EQUIV
COMB

PRED
PRED
COG
HCG
NOBG
NOTG
NO.

TEMP
TEMP
FLOW

RATIO
TEMP

EFFY
EMM
EMM
EMM
EMM
EMM


OF
OF
LBS/HR


OF


G/KG
G/KG
G/KG
G/KG
G/KG
301
"l
88
69
48.5
1
.2188
3444
0
.0000
0 .0000-
629.4393
0.5725
2.3267
"l.0000
302
"l
92
70
96 . 9
1
.4219
3172
0
.0000
0 .0000
1104.4730
"l.0000
0 . 3230
"l . 0000
303
1
74
68
98.5
1
. 2813
33 52
0
. 0000
0 .0000
424.4660
2.3409
0.5779
"l.0000
304
~1
87
70
141.5
1
. 3438
3275
0
. 0000
0.0000
950.9025
1.0133
0 .425?
"l.0000
305
1
100
67
124.2
1
. 3750
3241
0
.0000
0.0000
1030.7704
0.8641
0.4437
"l .0000
306
"l
100
68
139.8
1
.1719
3 503
0
.0000
0 . 0000
570.8785
6.0176
0 .6130
~1 .0000
307
"l
104
69
94 . 3
1
.1953
3483
0
. 0000
0 .0000
558.1266
~1 . 0000
0.3789
"l .0000
308
2
90
320
46.6
0
. 5469
2447
0
. 9 886
1 1 . 3665
1 .9862
"l.0000
0.2088
0.0979
309
2
98
350
47.5
0
. 5052
2285
0
.9794
20 . 5938
2.1579
0.0000
0.0898
0.0614
310
"l
94
310
48.4
0
. 3436
1562
0
.9100
90 . 0000
0.5925
0 .1868
0.0529
0.0529
311
2
385
350
49.4
0
. 5886
2820
0
.9964
3.6314
1.8390
0.0000
0.3384
0.3061
312
2
388
348
48.3
0
. 5098
2546
0
.9925
7.4545
0 .1425
0.0000
0.0164
0.0726
313
2
395
340
47.2
0
. 3945
2063
0
.9 522
47 .8285
0.1861
0 . 0000
0.0092
0.0336
314
2
340
4 50
88.2
0
. 5392
2603
0
. 9883
11 . 6792
~1 .0000
~1 .0000
0.0883
0.1590
315
2
364
4 50
93.1
0
.4726
23 57
0
.9727
27.2698
"l.0000
"l . 0000
"1.0000
0.0507
316
2
400
460
90 . 5
0
. 5043
2520
0
. 9849
15.0668
0.1441
0.0000
"l . 0000
0 .0332
317
2
435
460
90 .0
0
.4059
2085
0
.9216
78.3522
0.1806
0 . 00 00
0.0059
0 .0237
318
2
350
638
137.1
0
.4935
2408
0
.9665
33.4715
0.1474
0.0000
0.0378
0 . 0848
319
2
3 50
620
137.2
0
.4150
1976
0
. 8800
120.0000
0.3249
0 .0000
0.0000
0.0319
320
2
351
600
134.5
0
. 6484
29 81
0
.9928
7.1961
0 .1838
0 .0000
0 . 0673
0.1328
321
2
140
540
89 . 8
0
. 597 8
2659
0
. 9887
11.2516
0.3619
0 . 0000
0 . 0951
0.1367
322
2
115
490
90.0
0
.4857
2115
0
.9234
76.5982
0.5007
0.00 00
0.0074
0.0394
323
1
105
530
90.0
0
.4477
1922
0
. 8960
104.0000
5.1723
0.2758
"l . 0000
0 .0295
324
2
9 5
585
135.5
0
. 5847
25 55
0
.9769
23.0948
0.1235
0.0198
0 .0588
0.1144
325
2
105
550
137.3
0
. 5181
2229
0
.9286
71 .4471
0.9337
0.0117
0.0184
0 .0437
326
1
110
610
138.7
0
. 4796
2023
0
. 8900
110 .0000
4.5562
0.2561
0.0050
0.0399
327
2
412
7 7
48.2
0
.6017
2884
0
.9970
3.0281
0.4025
0.0286
0.1456
0 .2342
328
2
395
90
47.3
0
. 4641
2378
0
.9871
12.8502
0.4998
0.0000
0.0320
0 .0671
329
2
390
9 5
47.1
0
.3821
1968
0
.9244
75.6240
3.1701
0.0000
0.0000
0.0253
330
2
407
97
90 . 3
0
. 59 5 5
2853
0
.9939
6 . 0797
0.1211
0.0000
0.2646
0.3661
BURN=2	STABLE	PRED EMM = 1000*(1-EFFY)	NOB - BURNER DATA FOR NOX
BURN = 1	LIMIT	PRED EFFY FROM THEORY	NOT - NOX FROM VAPOR GENERATOR EX H.
BURN-0	GOING OUT
T able
VIII-29

-------
COMPARISON OF PREDICTED BUPKER INEFFICIENCY AND EXPERIMENTAL EMISSIONS DATA
PACE 12
RUN
BURN
AIR
FUEL
AIR

ECU IV
COMB

PRED
PRED
COG
HCG
NOBC
NO.

TEMP
TEMP
FLOW

RATIO
TEMP

EFFY
EMM
EMM
EMM
EMM


OF
OF
LBS/HR


OF


G/KG
G/KG
G/KG
G/KG
3 31
2
400
100
92.7
0
. 4834
2436
0
.9798
20.1670
0.1506
0.0000
0.0891
332
1
435
92
89.7
0
.4059
2086
0
.9223
77 .6826
358.7488
0.0000
0 .0000
333
2
392
95
135.2
0
. 4401
2160
0
. 93 66
83 . 4309
4.7839
0.0000
0.0000
334
1
395
1 00
140.3
0
. 3843
1908
0
.8800
1 20 .0000
"l.0000
0.0000
0 . 0000
33 5
2
92
590
91.3
0
. 55 56
24 57
0
.9 787
21.2564
0.1263
0.0000
0 .0770
336
1
90
630
92.6
0
.4667
1964
0
. 8900
110 . 0000
9.0811
0.0631
0.0000
337
2
90
610
4 8.5
0
. 5434
243 2
0
.9876
12.4134
0.1292
0 .0000
0.1511
338
2
90
660
49.6
0
. 5899
260 5
0
.9926
7.4003
0.1186
0.0000
0.0080
339
2
100
660
136.7
0
. 5685
2492
0
.9718
28.1970
~1.0000
0.1976
0.1838
340
2
395
90
13 8.3
0
.4997
24 78
0
.9739
26.0525
0.1455
0.0121
0.0239
341
2
1 50
100
9 2.4
0
. 5245
2378
0
.9725
27.4705
0.1383
0 .0000
0.0250
342
2
90
85
93.3
0
. 5472
2419
0
.97 53
24.7073
0.1283
1.0000
0.0000
343
2
102
1 00
91.3
0
. 5273
2344
0
.9 666
31.4349
0 . 1375
0 .0000
0.0452
3 44
2
100
1 00
91.9
0
. 46 88
1979
0
. 8900
110.0000
0.1555
0 .6505
0.0077
345
2
102
95
135.3
0
. 544 5
2386
0
.9606
39.3849
0.1330
0 . 0000
0.0240
346
2
105
95
13 8.2
0
. 4898
2051
0
. 8890
111.0000
1 .0000
0 .1853
0.0000
347
2
105
97
134.5
0
. 6523
2813
0
. 9882
11.78 54
"l . 0000
0 .0000
0.2097
348
2
107
95
90.1
0
.6022
2649
0
. 988 2
11 . 80 85
0 . 3708
0 . 0000
0.0275
349
2
1 00
1 00
91.9
0
.5137
2274
0
. 9592
40.7793
128.2549
?. 5352
0.0070
350
2
101
105
136.4
0
. 5697
2498
0
. 97 24
27.57 51
13.6585
0 .0998
0.0417
351
2
105
810
92.4
0
.6311
2749
0
.9906
9.3830
0.1105
0 . 0000
0.0973
352
1
110
815
93.1
0
. 5036
2230
0
.9510
48.9656
0 .1399
0.0000
0.0190
353
2
1 20
790
13 7.7
0
.6311
2749
0
.9861
13.8546
0.1105
1.3118
0.1497
354
2
430
630
81.4
0
. 5379
2672
0
.9915
8 .4997
0.1306
0 .0000
0.1018
3 55
2
438
610
82.0
0
.4309
2251
0
.9646
35.3630
0.1643
0 .0000
0.0112
3 56
2
4 50
625
4 3.8
0
. 5570
2762
0
.9964
3.58 51
0.1260
0.0000
0.1024
3 57
2
140
690
180.5
0
. 50 53
2111
0
.8800
120.0000
0 .1394
0.0000
0.0284
358
2
110
690
46.8
0
.4673
21 1 6
0
.9590
41.0407
0.1512
0.0000
0.1025
3 59
2
120
730
138.2
0
. 5022
2097
0
.8860
114.0000
0.1403
0 .0000
0.0666
360
2
380
570
92.6
0
. 53 22
2608
0
.9881
11.8535
0.1321
0.0000
0.1119
BURN = 2	STABLE	PRED EMM = 1 000 *(1-EFFY)	HOE - BURNER DATA FOR NOX
3URN = 1	LIMIT	PRED EFFY FROM THEORY	HOT - NOX FROM VAPOR GENERATOR EXH.
3VRN=0	GOING OUT
NOTG
EMM
G/KG
0.0891
0 . 4155
0.0573
0 . 040B
0 . 1287
0.0597
0.1019
0.0 565
0.1580
0.1076
0.0955
0 . 8349
0.0949
0.0690
0.1704
0 . 0952
0.4230
0 . 5474
0.1373
0.3315
0.1707
0 .013B
0.2070
0.1588
0.0432
0.1531
0.0779
0. 0795
0.1014
0.1649
Table
VIII-30

-------
COMPARISON OF PREDICTED BURNER INEFFICIENCY AND EXPERIMENTAL EMISSIONS DATA
PAGE 13
RUN BURN
/lltf
FUEL
AIR
E0.UIV
COMB
PRED
PRED
COG
HCG
NOBG
NOTG
NO.

TEMP
TEMP
FLOW
RA TIO
TEMP
EFFY
EMM
EMU
EMM
EMM
EMM


OF
OF
LBS/HR

OF


G/KG
G/KG
G/KG
C/KG
G/KG
361
2
400
6 20
90 .9
0.4673
2370
0
.97 50
24.9818
0.1512
0.0000
0.0154
0.0795
362
2
405
670
89 . 8
0.6305
2967
0
.9952
4.77 57
0 .1106
0.0000
0 .1986
0 . 3236
363
2
405
645
90.8
0.49 84
2501
0
.9840
16.0114
0.1414
0.0586
0.1774
0.2323
364
2
405
670
48.7
0.5691
2769
0
. 9960
4.0079
0.1232
0.4965
0.1836
0.1943
365
2
405
670
48.0
0.6244
2953
0
. 9974
2.6264
0.1118
0.0616
0.2725
0.1910
366
2
410
660
47.9
0.4309
2255
0
.9794
20.6335
0.1643
0.0000
0.0558
0.0648
367
2
405
670
137.0
0.5100
2530
0
.9782
21.7602
~1 . 0000
0 .0000
0 .0515
0.1270
368
2
410
690
137.1
0.4309
2087
0
. 88 5 5
114.4709
0 .1643
"l.0000
0.0279
0.0432
369
2
400
690
136.3
0.5691
2748
0
.988?
11.7979
0.2464
0.0000
0 .1502
0.2267
370
2
403
710
90.9
0.5475
2683
0
.9907
9.3231
0.3745
0.1464
0 .1043
"l.0000
371
2
426
720
137 .1
0.5738
2785
0
.9893
10.6911
0.9111
0.0000
0.1572
"l.0000
372
1
425
700
138.7
0 . 5438
2676
0
.9 8 56
14.3504
1 .1935
0.5236
0.0525
"l.0000
373
2
425
720
90.9
0.5325
2646
0
.9898
10.2252
0.1320
0.07 57
0.0984
~1.0000
374
1
430
700
90.3
0 .4913
2495
0
.9839
16.0845
5.8846
6.5336
0.0584
~1.0000
375
1
400
710
91.4
0 .4618
2346
0
.9725
27.4727
25.2456
108.5867
0 .0042
~1.0000
376
1
400
710
90.3
0.4711
2387
0
. 9766
23.3936
85 .4441
180.1626
0.0041
"l.0000
377

405
705
48.3
0.4 83 8
2462
0
.9903
9.7117
"l . 0000
0.7287
0.133 5
1.0000
378
1
428
730
4 5.5
0 .487 5
2496
0
.9919
8.0865
3.3900
5.0370
0.0687
~1 . 0000
379
1
37 5
720
47.0
0.4650
2364
0
. 9865
13.5247
8.5400
3.4403
0.0670
"l.0000
380
1
412
715
4 5.7
0 .4395
2294
0
. 98 3 2
16.8192
27.7000
75.2300
0.0055
1.0000
381
1
411
728
138.5
0.4 507
2262
0
.9438
56.1542
32.1684
66.5627
0.0037
"l.0000
382
1
411
721
138.7
0 .4501
2258
0
. 9429
57 . 1028
49.3421
156.2723
0.0053
"l . 0000
383
1
403
700
47.1
0.47 25
2418
0
.9890
11.0430
20.7706
57.4081
0.0710
1.0000
384
1
76
710
91.4
0 . 5775
2531
0
. 9830
16.9827
0 .4440
0.0488
0.0822
~1.0000
385

90
430
91.6
0.5625
2483
0
.98 04
19.6011
~1 . 0000
0 .0402
0.1436
~1 . 0000
386
1
91
470
89.8
0.5284
2339
0
.9686
3 1.3526
32 . 2044
8 . 5927
0.0271
~1 . 0000
387
1
80
740
CO
r-
3
0.5255
2353
0
.9 839
16.1210
4.0152
1.0001
0.0998
~1.0000
388
1
81
740
4 5.5
0.5337
2389
0
.9865
13.5103
53 . 5975
28.3174
0.0067
~1 . 0000
389
"l
8 5
390
48.2
1.2334
3423
0
.0000
0.0000
1194.6713
18.0370
0.5425
~1 . 0000
390
1
90
51 5
80.9
0.5400
2397
0
.9769
23.1205
0.1301
0.2117
0.2557
~i.ooon
391
2
93
550
80.5
0.4900
2144
0
.9 3 88
61.1581
77.7172
19.0275
0 .0098
1 • 0 0 0 0
BURN=2
- • -
STABLE

PRED Ei-'M
= 1 000
*(1 -EFFY)

NOB
BURNER DATA FOR NOX


BURN=1
	
LIMIT

PRED EFFY FROM
THEORY


NOT
NOX FROM
VAPOR GENERATOR EXH.

*URN = 0	GOING OUT
Table
VIII-31

-------
APPENDICES

-------
REFERENCES
1.	Frank-Kamenetskii, D.A., "Diffusion and Heat Exchange in Chemical
Kinetics", Quoted by Frank-Kamenetskii. Princeton University Press.
2.	Van't Hoff, "Etudes de Dynamique Chimique", Amsterdam, 1884.
3.	LeChatelier, Quoted by Frank-Kamenetskii after Jouget, "Mechanigue
des Explosifs", Paris 1937.
4.	Semenov, N.N., quoted by Frank-Kamenetskii, Z. Tnysik, chem. 48,
571, 1928.
5.	Vulis, L.A., "Thermal Regimes of Combustion", McGraw Hill, 1961.
6.	Dezubay, E.A., "Characteristics of Disc-Controlled Flame", Aero
Digest, July 19571	—————
7.	Scurlock, A.C., "Flame Stabilization and Propagation High
Velocity Gas Streams", Third Symposium on Combustion Flame and
Explosion Phenomena page 21, Williams and Wilkins, 1949.
8.	Longwell, J.P., and Weiss, M.A.., "Flame Stability in Bluff Body
Recirculation Zones", pp 98, Ind. Eng. Chem, s,itn!y, JL95J 8th
Symposium on Combustion.
9.	Zwick, E.B., & Bjerklie, J.W., "The Mechanism of Combustion
Stabilization in Monopropellant Reaction Chambers", Sundstrand-
Turbo, Pacoima, 1958.
10.	Clarke, A.E., Harrison, A.J., Odgers, J., "Combustion Stability
in a Spherical Combustor1*, pp 664, 7th Symposium on Combustion,
Buttorworths, 1959.
11.	Jeffs, R.A., "The Flame Stability and Heat Release Rates °f Some
Can-Type Combustion Chambers", 8th Symposium on Combustion
pp 1014, Williams & Wilkins, 1960.
12.	Williams,et al, "The Combustion of Methane in a Jet-Mixed
Reactor", 12th Symposium on Combustion, pp 913, 1969*
13.	Clarke, A.E., "Further Studies of Combustion Phenomena in a
Spherical Combustor", 8th Symposium on Combustion, pp 982,
Williams & Wilkins, 1960
14.	Herbert, M.V., "A Theoretical Analysis of Reaction Rate
Controlled Systems Part II", 8th Symposium on Combustion,
pp 970-981, Williams Wilkins, 1960.
15.	Hougen, O., Watson, K., Ragatz, "Chemical Process Principles",
J. Wiley & Sons, 1947.

-------
8c!orimeSric ~iorodeteraMon of [JiSro^on Liossde in fte fltmos^.ere
BERNARD E. SALTZMAN
Division of Spec/a/ Health Service, U. S. Department of Hwlth, Education, andWelftre, Cincinnati, Ohio
I he determination of nitrogen dioxide in the atmos-
phere hat heretofore been hampered hy difficulties ill
p.implc nleorplion and ln« k of specificity. A now spc-
nfic reagent has been developed nnd demonstrated to
obnorb efficiently 111 a nudget frilled bubbler ul levels
| below 1 p.p.m. The reagent is a mixture of aulfanihc
urid, /V-(l-nuplil h> l)-el liylenedinmino dihydroclilo-
riilr, mill nrclic ui id. A htnble ilm el color In produced
wilh 11 HctiRillvity of n fow pnrtR pi r billion for a 10*
iihii11lc sample lit I) 1 hit r per imnulc. O/ono in five-
fold excess nnd «»11io< ^ isea in t« nlold excels produce
only slight interfcm.fi cITVeiv, tin >c niny be reduced
furl her by means which are itohcriheii,
rOXIC oxides of nitrogen, liberated during the use of ex-
plosives, in welding operations, in the exhaust of internal
¦nihustion (ngincs, and in eliemieal processes involving nitration
III" urc of nitric aeid, are well known health hazards In
Tent ye irs new interest his been di reeled toward concentra-
nns of a few Lonllis of u j> irt per million of nitrogen dioxide,
¦ hibh are believed to pla> a vita! role >n the creation of irritating
|rtG (4. 10) ToMcologn studies (8, 0, 15, 32, £4) call attention
•	the fart th it mtro^r n dioxide ih the most toxic of Iho various
jlrogen oxul< h by a lar^o factor, and tlmt confusion in tho
initiation the health h \z \rds lus resulted from analytical
lethods wl< i fail to dilTerentmlo this o\idc from the others in
i mixture In terms of nitrogen dioxide, a figure of 5 p p m is
he maximum safe allow ibie concentration proposed (5, 9)
\I1 those considerations require its determination in air at much
mcr levels than previous^ though! necessary.
The major problem of past analytical methods has beea tho
'(faulty in absorbing the gas from a sufficiently large samplo
<*ulU have been uncertain for levels below 6 p p m Samples
iuRt be collected in large bottles for tho well known phenol-
Sulfonic acid method (3, G), nnd days arc requued for complete
'(sorption, low results hive boon reported (16) and confirmed
ji the present study bimilar difficulties occur with tho m-
.jlenol methods (11, £/>) Hoth dcteimino all nitrogen oxides in
lie fortn of nitrate, rather 111 m nitrogen dioxide specifically
Attempts h ivo boon made to urc reagents for nitrite icn, winch
«ouM hi specific for nilrogi n dioxide, but an absorption efficiency
/ only about .*>% was rej)oi ted (16) when a midget impmger was
wd Ilou'ver, these reagents were found to be very con-
ppieni foi Higher levels u^ing a glass eynnge for collecting tho
imple (i, iQ)t low levels havo been determined using a
ifh<» ' Continuous « unplcs have boon collected by using
If • it hipnd nir t< iii|m t itures (7) or alkali bubbleis {21),
¦ t.ii r being of unknown i (heicncy
present report deiK with the development and demon-
• in of a reagent wlmh is specific for nitrogen dioxide and
• l for continuous i impling with a high efficiency Tho
1'i.iblcin of 1 In mldprf Frilled hiihbl< r mid air l«
i«Mpl. d hi h rate oi 0-J hlei pei nuinito, it ficnxllivil y ul a fow
>iIh pi r inlliMii m nt t uiiM'd with u 10-immite samplo 'i he effect
•f vaxion* iid'TferhiK rough was found to l»o nlight.
Ai'i*>\n atus
Spectrophotometer, Iieeknnn ^lodol OU A set of matched
test tubes, 22 X 175 mm , giving an o,»tieal light pith of 2 02
cm was mod in a wpecial holder httcd to the Pjjertropholometcr
Midget Fritted Bubblers, all-ghs^, capacity GO nd , with up-
ward-facing, S-mm diameter fritted disks When used wnh
10 ml of the absorbing reagent, drawing air through at tho rate
of 0 4 liter per minuto should produce 20 to 30 ml. of fino froth
abnvo the solul ion,
Grab-Sample Bottles, having Htandard-Upor gmund-joint con-
nortion to nUjpeocka lor evacuation, with calibrated vohuncs
vaiying from JO to 200 ml Onhniry class-stoppered boro-
eihcatc glass bottles arc suitable 1 ifty-milhliter glass syringes
arc convenient for moderately high concentrations
RRAGI NTb
All rcigenls are made from analytical grado chemicals in ni-
trito-fiee w iter pi< pared bj' redistilling distilled water in an all-
glans still after adding a crystal caeh of pot/iasium perminganato
and of baiium hydroxide They are stable for Hovcral months
if kept well stoppered in brown bottles in the rofngerator
jV-(t-Naphthyl)-cthylenedjanime Dihydrochloride, 0 1%
Dis^olvo 0.1 grain of tho reagent in 100 ml. of water Stock
solution
Absorbing Reagent. Dissolve 5 grams of sulfanibc acid in
almost a liter of witcr containing 140 nd of glacial acctio acid,
add 20 ml of the 0 1% stock solution of A^-{L-naphthyI)-cthyl-
cnediaimuo dihydrochloride, ind dilute to 1 liter.
Standard Sodium Nitrite Solution, 0 0203 gram per btcr One
milliliter of this working solution produces a color equivalent
to th it of 10 pi of nitrogen dioxide (10 p p m in 1 liter of air
at 7li0 mm of mercury and 25° C ) Prepare fresh by dilution
from a stronger stock solution The latter may be prepared from
Merck reagent grade granular solid, which has been shown (17)
to assay 0(J 4%, in another study (5) it was found to assay
100 4%, dr> ing was said to be unticccssary» and a stock solution
of 8 grams per liter was found to be stable for 00 days
I'ltOCI DURE
Sampling for Levels of 1 P P M and Below. Place 10 ml of
absoilung re igcnt in a midget fritted bnbhli r and draw a sample
through it at the rate of 0 4 liter pet innmle until Milfiuent color
his developed (about 10 minute-*) \  connected to 1'ieev icuat^
bottle and the sample is collet toil I'm cock miiiuH the
volume of absorbing reagent Allow l.r> imuuU-a with oce-isional
shaking for coinph tc ahnoi ptimi uid eoloi -J'-velopmriit
Anothei more convenient but lesp nr. hi < k« | I m tin « • 1 t->r-
liige^, and 40 ml of air may be dnwn in I lie tune oi » ouphni;
If inHufheient color n e\peeled, the nbsm | >f ¦ >n m.»\ be • lipid id
Ijr fliJiluni' rln-irtinelv for 1 to 2	(• • uiiet wlni-l* il>. mi imir
hiiu»|Uilli>d uii] >o)dlllo|i|dfth til H ii in.
Dc tu mi unlioJl. Aft or colli • lion m mIihmi p| mn 1 he m miplo,
a ihrr» i nd vudnt rolor 			 < '«d».i .i.-v ¦ I i i im > <>ui.
plote within lo tumuUn nt oidmaiy	» 1 
-------
1650
ANALYTICAL CHEMlSThV
with standards vmunllv or read to u Biwctrophotonjetor at 6M)
dih, using unextN.H.l reagent as a relerenoo. Colon may hn
pirprivtiL if fti'M	ivilU only U to A% loi| In fthmifli*
iiliot) \H3t uav; lumaun, if uiiong oxidutiug or reducing gauuii me
present in Uio tuimplc in concentrations considerably oxcuediug
tli u of the nitrogen dioxido, the colore should be determined ao
Soon as powible to minimuo any loss
Standardization Add graduated umounts of Btandurd sodium
Dili itc sulution up to I ml. to a aei ich of 25-tnl. volumetric flasks,
and dilute to murks with absoihing reagent. Mix, allow 15
minutes for complete color development, and read the colors.
The 1-ml. standard is equivalent to J pi of mtrogon dioxide per
10 ml of absorbing reagent.
Calculations. For convenience, standard conditions are
taken as 700 mm of mercury and 25° C , thus only slight cor-
rettion is ordinarily required to Ret V, tho standard volume in
liter'* nf tho air sample quantities of mtrogon dioxido may bo
expressed n* microliter*, tA>, deiined as V times the paits por
million of nitrogen dioxido It lias been determined empirically
tlut 0 72 mole of sodium nitrite produces the same color as 1
mole of nitrogen dioxide, hence 2 03 7 of sodium nitrite is equiva-
lent to ljd of nitrogen dioxide.
Plot the absorb iiu **s of the standard colors, corrected for the
blank, against the milliliters of standard solution iieer's law is
followed Draw the straight line giving the best fit, and deter-
mine the value of milliliters of sodium mtiite intercepted at ab-
sorbance of exactly 1 This value multiplied by 4 gives the
stand irdizntion factor, i\f, dchncd as the number of microliters
of nitrogen dioxide required by 10 ml of absorbing reagent to
give an absorbance of 1. For 2-cni cells the value was 3 05
Then*
P p m of nitrogen dioxide = coircctcd absorbance X M/V
If the volume of the air sample, V, is a simple multiple of Mt
calculations are simplified Thus, for the M value of 3 65 pre-
viously cited, if exactly 3.65 liters of air are sampled through a
bubbler, the corrected absorbance ib also parts per million di-
rectly If other volumes of absorbing reagent are used, V is
taken as the volume of air sample per 10 ml. of reagent.
hXI'HtlMKYI AL
Preparation of Known Low Concentrations of Nitrogen Diox-
ide. The first Btep m the study wad the development of a suit-
able reagent which would give a high absorption efficiency with
continuous sampling, so that the low levels (below 1 p p m)
could be dotcrmmed These mtrogon dioxido concentrations
u ere prepared in the apparatus shown in Figure 1
Tho source of the nitrogen dioxide was a standardized uir
mixturo contained in a 40-hter carboy and available through an
aINgtuss system of 1-mm bore tubing und ground joints lightly
greased with silicone crease The mixture was made by intro-
ducing a few milliliters of nitric oxide, generated in a mtiomcter,
into the partially evacuated carboy, and flushing it m with air
until normal pressure was attained A few days u ere allowed for
air oxidation of the nitric oxide to nitrogen dio\ido and equilibra-
tion with the apparatus The resulting concentration of nitrogen
dioxide was 20 p p m , which was well within the range of ueeurute
analysis by eusting methods, and could bo determined by
collecting a sample in a GO-ml evacuated bottle through stopcocks
B und C The composition of tho air in the carboy wus found to
remain remarkably constant During a period of 4 months it
dropped to 15 p p m Most of this loss could bo accounted for
by the more than 100 )>ortions whi'h were withdrawn, each
amounting to about 1/I000th of the contents of tho rurboy The
vacuum that developed in the carboy was measuied and relieved
by udmitting outside air periodically, through operation of etoj>-
cotk D, which wus ordinarily kept In tho closed position
Known low concentrations of nitrogen dioxide were prepared
by accurate dilution of this standardized carboy uir mixture in
the following mnnnor. A 50-ml portion was withdrawn into a
glass syringe through stopcock A, and then slowly injected into a
1-hter-per-minute air stream by means of a motor-driven slide.
A dilution of 1 to 147 wna usually used, tho value could be varied
by moving the belt on the stepped pulleys of tho synchronous
motor (The second eyringo driven by the sumo slide, shown in
Figure 1, was used in later tosta to Inject an Interfering gas into tho
nir ntrMttti by ilmllar iiianlpplftiiun <•' sfcipoutiii ft )
The air stream used for dilution of the nitrogen dioxido wu
taken in through a universal typo gas-mask canister; tins re-
duced the normal nitrogen dioxide concentration in tho laborator)
air, which at times reached 0 1 p p m , to considerably less than
001 p p.m. (A U-tubo containing Asiante was found almost
equally efficacious.) A mixing chamber was provided for the
stream below each point of gas injection. Flow was controlled
by a critical orifice in the suction line to an aspirator in the hood,
preceded by a trap with a mercury manometer connection.
Figure 1* Apparatus for Preparing Known Low
Concentrations of Nitrogen Dioxide
1.	46-Liter earboy containing 20 p.p.rn. of nitrogen dloxltlo air mixture ¦
1	60-ML anmpling Itotllo
3.	Vacuum eoiineotlon to aspirator In hood
4.	CO-Ml rIuu lyrlnsoi omt molurnlrlvoa illde
5.	intoko for •icoml uu»
6.	Univcrvul gui mu«k canUtcr
7*	Sampling dovteo
B. Ilypuii device
9. Critical orlftio
Precise measurement of the amount of nitrogen dioxido injected
waa made by reading a counter on the motor drivo The appara-
tus was started and flushed with stopcock F in position to divert
the uir stream through a bypass of the same resistance aa the
sumphog devico At tho moment when a predetermined realms
was obtained, F was tui ned to direct the nir through the eaniph r
When tho syringe wus fully diuchar^od, a limit tm itrh stopped the
motor, and F wus aguii turned ut that instant to tho b)p>**
position again The volume of carboy uir mixture which liul
]>usscd through tho suinplitig dcvuc eould be c d< ulatcd from the
diltcrenco in counter readings und fi am nucromch r ine'isurctiu i»U
of tho synngo plungei diameter
A modified Shuw srrubber (14) was used ua the HainpluiK h-
vice for the scicening tests of vurious reagents, ulLhnugli a midget
fritted bubbler wus later found more efhuent 1 it tho Shuw
scrubber tho air stream entered a lift pump raising absorbing re-
agent to the top of u column, uhuh Imd u 10-tnm diarnileT
and was 00 nun high, packed with glass helices The siinipIc
was absorbed on the wetted surfaces of these helices .is the gu
and liquid both flowed downward, the uir then (lowed to il»
vacuum sourco while the absorbing lujuid drained b ick to the
pump. Twenty milliliters of absorbing reagent wire required
The absorption efficiency that could be obtained in the scrubber
at low concentrations of nitrogen dioxide was the critical factor 10
the tests. This cfheiem y vsua calculated u* follows
Absorption efficiency
AklbV,
where A9 is the absorbance of tho color obtained from tho scrubber

-------
JfOLUME 26, NO. 12, DECEMBER 19S4
! .t&gent (corrected for the value obtained in a blank run with no
jciirugen dioxide addition), A* ir the absorbanco of the color ob-
r tuned using the same reagent in tho evacuated bottle and
' amphng directly from the carboy (corrected for the blank value
of unexposed reagent), R, and ft* aro tho volumes of the reagent
ic chemicals, and of tho effect of various metnls added as
mtalyats Table I presents tho data whioh wcro obtained Tho
mgent finAlly adopted, hated as No 23, allowed tho highest
tflicicncy (77%) and excellent
color stability nnd sensitivity, 	
ilic maximal absorption of tho
red-violet color being at 550
1951
conditions, after which the pH is increased with & buffer aod tho
coupling ruagont is added for optimal color development. Tho
method, which is in accordance with studies of procedures for
nitrito (0, 18, IS), was used for Reagents 1 to 3 and 7 to 10
(Table I), as well as in other test** not shown with tjulfamlic acid
and sulfuric or hydrochlorio acid, followed by various buffers
and 1-nnphthylaminc. This method is subject to losses due to
decomposition of the unstable du/o intermediate dunng tho in-
tense aeration of sampling
Direct Coixjr Method, in which the reagent contnine all
ingredients and ufter absorption ptoduccs tho color with no fur-
ther operations This method is subject to Iobscs because of side
reactions between nitrite and tho coupling reagent, and because
of not having optimal pH for diazottzation and coupling. This
method, which was used for the remainder of the reagents listed
in Tablo I, was found to produce more color in the scrub-
ber, even though it produced less with a standard bottle sample
or nitrite portion. Tho direct color ty»e of reagent, contanuog
all ingredionts, was thorcforo adopted becauso of groatcr con*
vcuicnco and a higher absorption eificiomy.
Optimum concentrations nnd nciditiea for each combination of
chomicals wcro determined in order to obtain u true evaluation of
their worth The results in Tablo I showed that tho highest
Table 1
Screening Tests for Itcugrnts to Obtain High Absorption LfDclcncy
Absorbnnee
} Four combinations of chcmi-
j cals were tried:
!' The combination of sulfa-
mic acid and l-naphthylamiue
(Ucuficnts 1 to 0) was finally
rejected because of poor color
^ H ibility and occasional false
I color production after ucration.
I llctKcnt 4 is similar to but
I somewhat stronger than ono
, previously found oy l'utty (/ff)
I to give 5% efficiency in a mid-
get impinger but successfully
used in a 50-mI glass syringe
More stable and intense
colors were obtained with sul-
fa nilamide and jV-(l-Tiaph-
thyl)-othvlenediamine dihy-
drochloriae (Reagents 7 to 17),
with an eflicicncy as high as
64% (Reagent 12) These
chemicals were used in ponder
form with tartaric acid by
Jurohu (IS) and found to I to
ilored conveniently and satin-
factorily used in u 60-in I glass
syringe after dissolving in
water.
A higher efficiency wus ob-
tained by substituting anthra-
nilic acid for the sulfanilamide
(Reagent 18), since thiB was
known to have u very rapid
diuKotixation rate, but poor
color intensity and very blow
color development were found
liest results wore obtained
with tho previously unreported
combination of Hulfamhc acid
ami N-{ l-naphthyl)~cthy)cnc-
dmmino dihydrocMoridc (Ito.i-
gents ID to 32), which una tho
one finally adopted
Two methods of color de-
velopment wcro investigated
Stepwise Method, in which
pihi()l«'nliRorptioi» in thodi.ir.o-
b/in,' i ''iifcpnt is cur i m d out un-
der tiie optimal stioiiKly acid
No.
7
8
0
10
11
12
13
Absnrjitioo
Troccduro for Color	Ullirioocy,
Absorbing Roagent*	Dovelopmont ond Remarks*
D ~ Sulfandio acid, C "* 1-Naphthylamino, 620 mp
Btd sir
aomple*
8ld. nitrite
portion*
0 01 A'NaOH
0 05
0 05
w uav« D, 16% AoOH
0 05% D, 0 03% C. 14% AoOH
0 I2%D, 0 076% C 14% AeOH
0.6% D.O 03% C, 14% AcOH
Add 10% AcOH. 0 05% L>. after	30
20 mm add 0 03% C
After 10 mm add 001% C	27
Alter 10 mm add0 03%C	40
Direct color	62-00
Direct color	OA
Direct color	66
0 188
0 183
o m
0 184
D — Sulfanilamide, C •» #-(I-naphthy!)-«tliylsncdtamino dihydroeblorida. 644 m»»
0 02% P, 1% 1ICI
0 02% D, 10% AeOH
0 02% D. 10% AcOH 2 6%
II«bO«
0 6% D, 1% HC1
0 02% D, 0 002% C. 1% HC1
0 6% D.O 002%C, 1% HCI
4 0%_ D, 0 003% C, 1% HCI,
4o% Ethylene glyool
1% HCI.
Eth '
4 0% D, JO 004% C,
l6% AcOlI. 8% Ethylene
glycol
Same as 12 + 0 0026% Cu(II)
Same as 12 + 0 05% Fe(II)
0 6% D. 0 002% C, 2 8% H»PO<.
+ 0 00% Fe(l L)
After 0 mm add 0 002% C
Alter 15 nun odd 0002% C
After 16 nnn add 0 00i% C
Vory slow color development
After 15 inin add 0 002% C
Direct color
Direct color
Direct color Ethyleno glycol
added to increase •olubility
ofD
Direct color Ethylene glycol
and AcOH added to incrouo
solubility of 1)
Direct color Test of catalytic
effect
Dirret color Test of catnlytlo
cfTect
Dircrt color Test of cotalytie
cfTect
6A
40
40
62
48
04
0 100
0 184
0 320
0 080
D Antliramlic acid. C « N (l-nnplithyl) ctliylenedmmino dihydrnchloridc, 600 m*
0 06% D, 0 002% C, 0.1% HCI Dim t color \«ryslon culvr	SJ	0 140
development
D ~ Sulfanilio acid. C ~ Af-(l-n i|>litliyl)-cthylcncdiAMiinc dihydroehlorido 650 ni*
19
20
21
22
23
24
26
20
20
30
31
33
0 05% D, 0 002% C 14% AcOH
0 fi% D.O 002% C. 6% HCI
				% tartaric
0 6% D, 0 002% C, *
acid
0 4% D.O 025% C, 14% AcOH
0 6% D.O 002% C, 14% AcOH
8arae as 23 + 0 005% Zn(Il)
Game as 23 + 0 125% Zn(II)
Same as 23 + 0 005% Fe(II)
8amo as 23 + 0 05% Fo(lJ)
8amo as 23 + 2 5% re(II)
Direct color
Direct color
Di r< ct color
Slow coupling
Same as 21 + 0 01% Fe(IU)
Same as 21 + 0 005% V(V)
Bamo as 2J + 0 006% Co(ll)
Same as 23 + 0 00% As(IH)
Direct color
Direct color
Direct color
effect
Dirri t color
elTeet
Direct color
cflrct
Direct color
cdcct Unstable
obtained
Dmotoolor It At of catnlj tic
cfT' rt Unatablo cuior« ob
tamed
Turbid brown color developed
No color dovoloprd C dcitrov d
Dirret color lilr Clip t> in of mti < u xulo
CMi tlnuiilo (yr CtO ml ul JO p y in ) absorb* d in 10 ml of reagent
. >Ul Modified Hhaw scniMirr usrd at 1 ht^r r« r i-«
* Calculated for a standard sample of Ipl of m
U an evacuated bottle
i Calculated for 1 74 v of polasftJum ni^iln (unul'i he cqntvnli nl l
		iuTvi	
		 ....			I micrulitcr of mtruuui dioaoio if 0 ft mote
of poUsjUum oltrits were equivalent to I nioln of nitrugon diondo) in 10 ml of roncoDt

-------
ANALYTICAL CHEMISTRY
Table II.
84	0 02% D, 14% AoOR
35	0 $% D, 14% AeOH
SO	0 6% D.Oo63%C
23	0.6% D, 0 002% C. 14% AeOH
37	0 5% D, 0 002% C, 60% AoOH
fritted
fritted
frittad
possible concentration of di-
nzotismg reagent waa desir-
able, not only waa the absorp-
tion efficiency increased, but
even the color obtained in a
bottle v itli a standard air
samplo (Reagents 11 and 12)
Too high ft concentration of
the coupling reagent, on the
other lmnd, reduced the color
produced (Reagents 22 and
'2\), probably bccauso of in*
creased side reaction directly
between this reagent and the
nitrite. A high acidity (Rea-
gent 20) greatly slowed the
coupling step for the finally
adopted combination of cheini-
tula Acettc acid was best he-
< uusc it provided the best com-
promise pll and also had sur-
face tension properties which
provided a fine froth in the
biraphng device Reagent 23,
based on these principles, wus
found to give the highest ab-
sorption efficiency.
The effect of various metuls
added as catal>sts wus slight
(lteugents 15 to 17, 24 to 32). The most effective metal was
0 05% iron(II) (Reagent 27), which improved absorption and
color intensity, but was considered undesirable because of color
instahiht) which would result if oxidation to the iron(III) form
occurred (Reagent 29)
Nitrite Equivalent of Nitrogen Dioxide. Practically, stand-
unlirution of tho rcngnit is best achieved with stundurd nitrito
solution, rather than with difhcultly prepared standard gus
samples The initial presumption was that 0 5 mole of nitrite
would be equivalent to 1 mole of nitrogen dioxide, by dissolution
in water of the latter to give equal quantities of nitric and nitrous
acids (Equation 1 below) The last two columns of Table I,
giving the absorbances obtained with 1 pi of nitrogen di-
oxide in an air sample, and with the equivalent amount of
nitrite on the above basis, showed that this presumption wus
not correct, dividing the first figure by twice the second gives
the actual molar eqmvilcnt obtained The previously men-
tioned study by Patty (16) found a relationship of 0 57, al-
though a satisfactory explanation of the difference from 0 5
was not presented In an effort to find the cause of disagree-
ment, a more complete invalidation was undertaken of tho
rHntionnhip between the color obluincd iu nn avueuuted bottle
with a stundurdized air sample, mid the color obtained in solu-
tion with standard nitrito reagent
The effect on the color intensity, which could be produced
by varying the concentrations and combinations of the ingre-
dients of tho final rcngent, in shown in Tablo II All solu-
tions gavo ubout the mmim iclor intuinty with a standard
nitrite portion, but thf color intensity with a atundurd nir
wimple varied more widely Values close to 0 5A1 equiva-
lence were obtained when the air sample was ubsorbed in
acetic acid alone (Reagent 33) or in a dilute sulfanihc-acetie
acid reagent (Rcngent 34) A value of 0 514/ equivalence
was obtained with Reagent 23 (the finally adopted reagent)
in another test not shown m which the air sample was increased
to 500 *il (of nitrogen dioxide) In this cose the range of the
reagent was exceeded and only a weak orange-red color was ob-
tuined, but upon dilution 100 times with additional reagent
the characteristic color was obtained However, higher values
Mere obtained with stronger sulfandio acid (Reagent 35), and
Influence of Reagent Composition on the Nitrite Equivalent of
Nitrogen Dioxide
Absorbaooe
8id air Sid oitrite
sample* portion*
Procoduro for Color
Devalopiacnt and Remark**
Absorbing Reagent*
D - SuUaoilio aold, C " Ar-(l*naphthyl)^thylenedlamlne dihydroeLloride, 560 mji
14% AeOH
Molei of Nitrite
Equivalent to
, Mole of N(V
Absorb 20 mla , add 0 02% D,
after 16 win., add 0 002% C
Absorb 20 mm , add 0 003% C
Absorb 20 mia , add 0 002% C
Abaorb 20 min Direct color
Absorb 20 una Dlroet color.
Final roagent
Abaorb 20 min Direct eolur
0 183
0
180
0
48
0 100
0
1HU
0
n
0 250
0
lul
0

also be written to show two molecules of sulfamic acid combining
with tho nitrogen tctroxide, with the name end result
Absorption Efficiency with Various Sampling Devices. Mt"
tho most suitable reagent hud been developed in conjunction
with the modified Shaw scrubber, it was found that much better
efficiency could be obtained using midget fritted bubblers
The acetic acid content of the reagent mndo possible a fine sod
stable foam of 20- to 30-ml volume abovo 10 ml of reagent nnd

-------
.OLUME 2 6, MO 12, DECEMBER 1954
1953
>)) Gray, 1C LHi, MncNuinoo, J K , and Gnldltorfi, 3 B , Arch,
Ind Ilyg and Otcu/xUxuncl Mid , 6, 20 (1052)
',0) Hiagon-Snnt, A J,/«W Khq Chen , 44,1342 (1052)
1)	Holler, A C , and IIu< h, R V , Anal Cuem , 21, 1385 (ILK1))
2)	Jacobs, M 13, "1 ho Analytical Chomistry of Industrial Poisons,
Hazards, und Solvcnt-s," 2nd od, p 358, New York, Inter-
scicnco Publishers, 1049
3)	Johnston, It S, and Yost, D MChem Phy* , 17,386 (1640).
j) hicwlb&ch, H, lsi» JCno Cuem , Anal Ed , 16, 700 (1044)
j) lalownlcy, I. W , tt al, J Ind Hj/q Toxicol, 23, 129-47
(JU41)	*
ib) rally, F. A andlVtty.G M ,lbid , 25,301 (I'll t)
I) Kcindollar, W 1 Inu Lng Cuem.Anai hi>, 12, 325 (1040).
,S) Kidcr, B V , with Mellon, M G , Ibid , 18, 00 (1040)
(10) Slitnn.M H ,!hut, 13,33(1011)
(20)	fcjlinidmun, L , and Yuaw, J y , Am <7n« At*oc. Proe , 24 (IM2),
277
(21)	Stanford Research Institute. "Third Interim Report on tha
Sn>ot? Problem in IjOs Angclos County," 1050
(22)XU	S Public Health Sem
-------
.OLUME 26, NO 12, DECEMBER 1954
!
-ovidcd a large surface area for good absorption. An upward*
, .nng fritted disk was better than vertical or downward-facing
because there was loss coalescing of bubblc9 with conse-
urnt loss of surface area In Table III arc shown the atoorp*
jin efficiencies obtained for various h unphng devices The threo
urtget fritted bubblers te^trd showed 01 to 09% efficiency At
»i to 0 1 p.p in of nil r i^m dioxide 13y using two in series
' jiocrs«ary, practicall) 100% efficiency may be obtained, a tost
j,ili «ii« b arrangement showed 1 he second bubbler recovered 01%
. tin lew hundn »lt-is a part per million which passed the
i bubbler. The fritted bubbler with thr Inchest pressure drop
^ showed the highest efficiency.
Standardization against Known Concentrations of Nitrogen
oxnlc. In tho work described above, the assumption was
i*b* tluit the fin.tl reagent gave true values when used in the
-wonted botllr to Simple undiluted carboy uir mixture, and
t.-nrption cfTicientiiH were calmlutixl on the basis of the relative
nlor obtained in this manner as compared to that obtained
l.in sampling diluted carboy air mixture It remained to
jrnionatr.vtc tbr arnuacy of this presumption by sampling higher
; (oiMcntratioofl of uitrogco dioxide of known value. Three
, .\-
'¦vi'd valuo of 20 p p m in the carboy of Figure 1 was compared
Hi the value exported from the amount of nitrie oxide intro-
duced from the nitrometer The following two systems, how-
ever, yielded good agreement between calculated and obtained
VjIllCM
In the second system, 0 4 grim of liquid nitrogen dioxide was
nmiritel) weighed in a scaled glass ampoule, which was then
broken in a closed, stainless-steel, cubical chamber which meas-
ured 5 feet on edge Ad electric fan was used to mix the con*
tents The theoretical concentration was computed at 06 4
p p ni The following results were obtained;
Minnies i/tcrbreoking 10 20 30 36 00 126 166
P P M . prwentmothod 00 7 00 6	63 8 09 1 01 7
PPM, phenoldisulfonic
toid method	04 3 OA 9
The concentration dropped slowly, but the data indicate that
the early samples were substantially identical This run was
uAed as the absolute standard of the entire investigation The
analyses showed good agreement, although samples of different
size were taken
In the third system known concentrations of nitrogen dioxide
nere prepared by means of four flowmeters A special flowmeter
was constructed to measure very small tank nitrogen dioxide
flown below 10 ml per minute The principle of this flowmeter
wn» that the gas was made to flow through a fino fritted*gluss
filter and a Cue capillary tulrc, mid tho pressure drop was meas-
ured by a manometer containing fluorocarbon liquid, it was
calibrated by pawing tho gas directly into weighing bottles con-
fining- Aecarite and noting tho gain to weight, manometer
readings, the (low time This flow was injected into a metered
air stream, and a portion of the mixture was taken off through
an all-glass rotoraeter into a second metered air stream. The
final mixture flowed into a glass jar from which samples could bo
19S3
collected. It was necessary to scrub all tho air with dilute
dichromato-sulfunc acid to romovc impurities ouch as ammonia
which precipitated or consumed nitrogen dioxide.
The results of the analyses of simultaneous samples were as
follows
	P P M	
Flow meter value	S A 16 I 27 3 40 6
Aoalvnisby prosont mnthnd 8 2 10 6 27 8 43 0
Annlj eta by pbenoldiaulfnoie
aeid method	5 3 10 0 20 6 33 4
Com! agreement was obt umd with the present method even
with sample fi&ct varying fmm 45 to 'JSO ml
The phenoldisulfonic acid procedure w/w systematically low,
as haH boon re]>ortcd for smulir hi is concentrations (1G) Pre-
vious tests with thiR proceduie hud shown Unit ubworptionof lower
concentrations in large bottles was very slow, although 3 days
had been allowed, slightly higher results could be obtained with
l-«eek absorption Thew «nmplcs were collected in 2.5-hter
acid bottles with 15 ml. of absorbing reagent, in the figures
previously quoted for the stainless-steel chamber, 600-ml bottles
wcic used because of the higher concentrations and 1 day of
absorption in tho refngerntor uiis adequate.
The result of those studies w.is the absolute standardization of
the method and the establishment of tho validity of the absorp-
tion cfliucnciex which weie ohtaimd with the apparatus shouo
in Figure 1
Y V FF<*T OK INTI IIFI.IIIM; GAbKS
The effect of various inleif. ring gosos wm found to be unim-
portant unless tho coneentr ition w.n mucli higher thAn that of
the nitrogen dtuude Because of the possibility of widely varying
sample sizes and concentration1*, all the results below are ex-
pressed in terms of microliters (corrected volume in liters tiroes
parts per million) per 10 ml of absorbing reagent For compari-
son, 2 or 3 /il of nitrogen dioxide arc ordinarily required to de>
vclop a color of suitable intensity, the normal color was found to
fade at the rate of 3 to 4% of il»«orbance per day
Ozone *Thc effect of oionc is 
-------
1954
ANALYTICAL CHEMISTRY
oxidation of nitrogen dioxide by the ozone. The correction to
the analysis for this effect was roughly computed as +10% for
1 p p m. of ozone, +21% for 2 p.p.m., and +60% for 5 p.p m ,
tho log of the correction factor woe proportional to tho concen-
tration of ozone This method waa very convenient and satis-
f ti lorv for thm 2 or 3 p p m of ozone; at higher valued tho
toi i in liuii bccime Ingh and uncertain.
Onhnary, rc igent-grade manganese dio\i!furcncc from other nitro-
gen oxides is negligible
The evaluation of the interference of nitric oxide, NO, is cook
plicated by the fact that this comjiound is slowly converted by'
air to nitrogen dioxide However it has been studied in the ab»
Bence of air in tho gas industry (£0), using snlf inihe acid aod l-
naphthyhimmc, mid fouud not to produce tuiv eolor unless
converted to nitrogon dioxide by a apocml oxidizing nrrublxr.
Since tho present reagent produces a color by a gimil ir reaction,
it may safely be said thut this gas does not interfere
Equilibrium calculations show that nitrous acid anhydride,
NsO), and nitrogen tetroxide, NjO,, do not exist at concentrations
of 100 ppm and below Kinetic data show that their dissocia-
tion ia practically instantaneous Hence these nitrogen oxtda
may be disregarded
Nitrogen pentoxidc is rarely found, because it is readily hy-
drated to nitric acid vapor, and is also an unstable compound
which is very sensitive to heat, the half life is 6 hours at 25° C,
83 minutes at 35® C , and only 6 seconds at 100° C. The de-
composition products are nitrogen dioxide and oxygen Tbw
compound was prepared by mixing a stream of nitrogen dioxide
with ozone in 0 5 p p m excess using the flowmeter apparatus
previously referred to The stream contained 25 p p in of
nitrogen pentoxidc (equivalent to 50 p p.m as nitrogen dioxide),
and gave a test for about 5 ppm, of nitrogen dioxide It u
likely that this *as due to impurity or decomposition of the nitro-
gen pentoxidc
Nitric acid docs not interfere with the determination When
added in solution to tho reagent it producod no eolor, nor did it
affect the development of color with nitrite solution or nitrogen
dioxide gas In tho form of vapor, u 5000-p p m sample col-
lected in an evacuated bottle gave a test for only 23 p p m. of
nitrogen dioxide The sample was prejjared by allowing a small
amount of concentrated nitric aetd to atand in a closed bottle,
with the addition of a crvsUl of sulfamic acid to destroy nitrous
acid impurities The small interference found may actually be
nitrogen dioxide produced by decomposition in spite of this pre-
caution
Other Interfering Gases. A number of other gases were in-
vustig.iU'd hv adding them in the form of water solution to a
reagent solution whiJi contained a color equivalent to about 2
pi of nitrogen dioxide The nmouut added was equivalent to
125 pi of interfering material Hydrogen sulfide produced no
efTcct Chlorine partially bleached the eolor instantly, lausioga
45% lo«w and changing the tint to orange, the final color remained
perfectly stable Hydrogen peroxide increiiscd the color slightly
(+4% in 2 bouts), after 3 days the color h.ul increased 10%
and h vd a slightly difTercnt tint with lets violet than tho normJ
color Formaldehyde produced no appreciable efTcct in 2 hour?,
in 3 days a 15% greater than normal color loss occurred u ith
production of an oiange-yellow tint In the presenee of 1%
acetone (lined forsnlfui dioxide) tho iut< rfer» tui-iofid! (lie-** tiu-
tcriulri was the tJ'ime, except lor tint of foini i-idelnde wlmh
still did not interfere within 2 bourn, hut caused dnio«t enmj-1' '•
Iokh of color in .1 dnj s
ACKNOW!.MM,MI\Nr
The author ih grateful to J T Mountain for many helpf'il
suggestions, and to D II livers and 11 E Stolangcr, under
whoso direction tlic wurlc wius carried out, for their valuable
review and criticism
11 ri- ItAl I'Uh i I I fcl>
(1)	Avorull. P H . Hart. W F . Woodbury. N T . and Hi l,y. W.
R , Anal Cubm , 19, 1010 (1*117)
(2)	Humes, If, and Folkard, A K . Amilytt, 76, 500 (1951).
(3)	Heutty, R L , Bargcr, L B , and Schrenk, H II. U. S Bur
Mines. Rcpt Invest 3687(1043)
(4)	lilacct. F E.Ind EnU Chan . 44. 1130 (1952).
(f<) nriujUHj.lt C . Anal Chem.23, • >(1051)
(0) Cholnk. J, and McNary, II. J hid, UyQ Tox\col, 25, 354
(1013)
(7) Kdrar, J L., and Pnnrlh, F \ , J Chrm Sue . 1941, 511, 519.
(5)	J I kind, 1 f II , J Ind UtiU Toxical , 28, 37 (HMO)

-------
OX8D5S OF N3TROGEN
GRK5S-SAITZMAN f.lEFTlIOD
A PCD 11-56
SCOPE
This method Is used to determine small concentrations of nitrogen dioxide fevmd In the
atmosphere. Tho lower limit of the method Is about 0.1 o.P.m. In a 5C9-ml. samplo
bottb. By altering the volume of the sample, nitrogen diox'do In specific soirees
can also be measured. However, the phen^dlsuLfeols geld method Is generally em-
ployed by thl; laboratory for source tasting. (N.. B.)
Hie ::t;':s are collected In evacuated bottles containing the absorbing wluf'on.
The aktcrbln; solution consists of a mixture of sulfanlllc acid, acetic ac c, c- J N-
(l-naphthyl)-ethylenedlamino dihydrochloride. After shaking, tho nitrogen dioxide
dlazotizes tho sulfanflic acid which then couples with N-(l nofhthyl)-«thylanedio-
mine Forming a dye. The intensity of the color Is measured with a colorimeter and tlio
concentration of nitregon dioxide read from a calibration curve.
Chaney rotary sampler (Figure 1), 500-mI. bottles with narrow necks containirg
tabulated sidcarms, 3-inch pieces of heavy-wall gum-rubber tubing, scrcv
clamps, solid-glesj plugs, and serological stoppers(Figure 2).
ANALYTICAL:
Spectrophotometer (Coleman Universal Model 14), microcuvettos (Colemon No.
14—315, minimum volume 2.5 ml.).
REAGENTS
COLLECTION:
0.1% N-(l-Naphthyl)-{thylenedi3mlne Dihydrochloride. Dissolve 0.1 gramof
in 800 ml. of water. Add 140 ml. of glacial acetic acid and 20 ml. of 0.1%
N-0-naphihyl)-ethyloned:cmine dihydrochloride solution. Dilute to I liter.
ANALYTICAL:
Standar£JodiunJ^itrifs_So^ Accurately wolgh 0.2755 gram of sodium
N. B.: This laboratory method lias been superseded by the District slnco August 1956
for air moVcring purposes by an automatic recording instrument. Nitrogen Oxides
Recorder, Model 3011, manufactured by Eorman Engineering, Inc., Nor.,'i Hollywood,
California, to District specifications and later modified by the Lo; Angeles County,
Air Pollution Control District.
METHOD SU.V-MARY •
SPECIAL APPARATUS
COLLECTION :
APCD 11-56
P. 1 of 6

-------
nitrite (NaNOjr 98% weight assay) and dissolve In water. Make up to volume
In a 1 -liter wlumefrlc flask. Transfer 1 ml.of the solution, by means of o cali-
brated plpet, to a 50-fnl. volumetric flask and moke up to volume with water.
One milliliter of the latter solution contains the equivalent of 5 . of nitrogen
dioxide gas (based on the 0.72 factor from Soltzman) and will be referred to,
hereafter, as the standard solution.
Tronsfer the followingamountsof the standard solution to a series of 25-ml. vol-
umetric flasks using a measuring plpet or 5-ml buret 0.1 ml. to the first, 0.2
ml. to the second, 0.3 ml. to the third, 0.4 ml. to the fourth, 0 5 ml. to the
fifth, 0.6 ml. to the sixth, 0.7 ml. to the seventh, 0.8 ml. to the eighth, 0.9
ml, to the ninth, 1.0 ml. to the tenth, 1.1 ml to the eleventh, 1.2 ml. to the
twelfth, 1.3 ml. to the thirteenth, and 1.4 ml. to the fourteenth Dilute each
solution to volume with absorbing reagent. The flasks will then contain the
oqulvolrnt of the following concentrations of nitrogen dioxide gas in micro-
grams per 10 ml. of absorbing reogent, respectively: 0.2, 0 * 0.6,0.6, 1.0,
1.2, 1.4, 1.6, 1.8, 2.0, 2.2, 2 4, 2.6, and 2.8. Shake the solutions and
allow iwaiwiufwi i j minutes for complete color development. Place the samples
In the spectrophotometer cell and read the obsorbance of the solutions at 550
mu against a blank of water. Obtain a blank value by reading the absorbonce
of an unexposed sample of the same botch of absorbing solution. Subtract the
blank value from the sample values. Plot the corrected obsorbances vs. micro-
grams of nitrogen dioxide per 10 ml. of reagent on rectangular^oordinote graph
paper.
IN (approximate) Sodium Hydroxide. Dissolve 4 grorrs of sodium hydroxide
pellets in 100 ml of water.
0.001N (approximate) Hydrochloric Acid. Dilute 1 ml. of concentrated hydro-
chloric odd to 100 ml. with water and then dilute 1 ml. of this solution to 1
liter with water.
COLLECTION OF THE SAMPLE
Soak the serological stoppers and the 3-inch lengths of rubber tubing in IN sodium
hydroxide overnight. Rinse with distilled water, then with 0.001N hydrochloric acid,
and again with distilled wator, Allow to dry.
Piece o serological stopper on the neck of the collection bottle. Attceh a piece of
rubber tubing to the sidearm of the bottle and evacuate to a pressure of about 25 mm
of mercury. Tighten the ccrew clamp cn the rubbe- ^ z- ' '".connect From the
source of vacuum. Insert the solld*glass plug Into the end of the rubbor tubing. Add
10 ml. ofabsorblng solution to the bottle by poking the needle of a syringe (contain -
ing the absorbing solution) through tfe serological stopper. The vacuum of the bottle
will drew the obsorblng solution in. Remove the syringe. Number the bottle in any
convenient monner. Load the bottle on the circular platform of the Chaney rotary
sampler. The top of tho bottle fits into o sleeve on o guide plate which has the time
APCD 11-56
P. 4 of 6

-------
of the sample marl-ed on it, and whu-h is on tho same shaft as the platform. The cir-
cular platform rotates by means of a timing mechanism so that a new bottle Is placod
into positron for sampling each hour. The sampler holds 8 bottles.
The sample is taken by means of a hypodermic needle connected to a piece of tubing
leading to the otmosphere. When o bottle moves underneath the needle, the guide
plate activatesamicroswitchmountedon the solenoid assembly. The solenoid is ener-
gircd and the plunger of the solenoid moves the hypodermic needle down through the
stopper on the bottle. After about 30 seconds, another micros witch on the solenoid
assembly loloasos the.rolenoid. This procedure is repeated for each sample bottlo.
An olochic interval timer turns the power to tho sampler on and off at any preset time.
Upon arriving at the sampler location, turn the power switch and drive switch off.
Remove the previous samples, and record the numbers of the bottles, times, and date .
Place tho next set of bonles in position on the platform and note tho number:, of tho
bottles, times, and date. Check to see that the needle will hit the stopper, and not
rhc guide plates, by fuming the platform by hand slowly until the first microswitch
c lic-o Manually depress the lever arm holding the needle slowly to within about 1/4
uich ol the stappci (do not puncture the jfoppu). Jic needle must bo approximate!)
in the center of the stopper. If it hits ony of the metal parts, it must be adjusted or
bent, so that it is in the correct position. This procedure is repeated for each bottle.
Generally, if the needle is aligned for the fin, t bottle, it v»ill bo correct for tho others.
Manually turn the platform to the correct time, check the timer for correctness, and
turn the switches on.
If if is desirod to take a sample manually, the sample bottle may be punctured with a
needle, or the glass plug removed and the screw clomp opened for about 10 seconds
?-hter sample flask, sirrl Jar to the ones used in the Phenoldisulfonic Acid Method for
nitrogen oxides, may also be used. In this case, place 10 ml of absorbing solution in
the flask and evacuate to the /apor pressure of the solution. Close tho screw clamp,
™d insert the solid-glass plug in the rubber tubing. Take the somple by opening the
flask for about 10 seconds Replace the screw clump, solid-glass plug, and return to
the laboratory for analysis.
SAMPLE PREPARATION
Shake the boltle (or flask) containing thesample for 15 minutes on a mechanical shakrr
to allow for complete color development
ANALYTICAL PROCEDURE
Transfer the sample from the bottle (or flask) directly to the spectrophotometer cell
and read the absorbance at 550 m /j. against water. Obtain o blank value by reading
the absorbance of the origtnai absorbing solution. Subtract the blank value from the
sample value to obtain tho corrected absorbanco. Read tho weight of nitrogen dioxide
corresponding to the corrected absorbance from the calibration curve.
APCD 11-56
P. 5 of 6

-------
ADDITIONAL NOTES
Whan operating theColeman spectrophotometer, check the setting for zero absorbance
bo for© eoch reodlng since the Instrument drifts gradually. To simplify this process,
one of the two mlcrocuvettes In fhe holder is olwoys kept filled with water ond the
zero checked against this cell before eoch reading. Since it is Important that the por-
tion of the cuvette in the light path be completely full, the analyst should realize that
the cuvette Is tilted slightly ond should fill It so that the side arms are about half full.
If only a few samples are to be analyzed, the spectrophotometer may be zeroed against
the blank. However, if a large number of samples are to be analyzed. It is better to
zero the Instrument with water and subtract the blank. This will avoid the possibility
of low somple values due to the gradual absorption of nitrogen dioxide from the labor-
atory otmosphcre by the blank.
REPORTING AND CALCULATIONS
Calculate the parts per million of nitrogen dioxide as follows.
0.532 W	(1)
vc
where
W = micrograms of nitrogen dioxide per 10 ml, of absorbing solution
vc - volume of air sampled, liters, at 760 mm. of mercury and 25°C .
Generally pressure and temperature corrections ore neglected,
and the measured volume of the bottle or flask is used
To convert to grains per standard cubic foot (60°F. and 1 atmosphere), multiply the
parts per million by 8.48 X 10"^,
REFERENCES
Grless, P., Ber„ 12, 427(1879).
Saltzmon, Bernard, Anal Chem , 26, 1949-5<>{1954).
APCD 11-56
P. 6 of 6

-------