APTD-1498
January 1973
NITRIC-OXIDE MEASUREMENT
IN A SIMULATED
SPARK-IGNITION ENGINE
II.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Air and Wa§le Management
Office of Mobile Source Air Pollution Control
Emission Control Technology Division
Ann Arbor, Michigan 48105
-------
APTD-1498
NITRIC-OXIDE MEASUREMENT
IN A SIMULATED
SPARK-IGNITION ENGINE
by
Robert F. McAUivy. Ill
Richard B. Cole
Combustion Laboratory
Department of Mechanical Engineering
Stevens Institute of Technology
Hobokcn, New Jersey 07030
Grant No. R-801874
(Formerly 5R01 AP00847-03)
EPA Project Officer: Curtis E. Fett
Prepared for
ENVIRONMENTAL PROTECTION AGENCY
Office of Air and Waste Management
Office of Mobile Source Air Pollution Control
Emission Control Technology Division
Ann Arbor, Michigan 48105
January 1973
-------
This report is issued by the Office of Mobile Source Air Pollution Control,
Office of Air and Waste Management, Environmental Protection Agency,to
report technical data of interest to a limited number of readers. Copies
of this report are available free of charge to Federal, employees, current
contractors and grantees, and nonprofit organizations - as supplies
permit - from the Air Pollution Technical Information Center, Environ-
mental Protection Agency, Research Triangle Park, North Carolina 27711,
or may be obtained, for a nominal cost, from the National Technical
Information Service, 5285 Port Royal Road, Springfield, Virginia 22161.
This report was furnished to the Environmental Protection Agency
by Stevens Institute of Technology , Hoboken, New Jersey 07030, in
fulfillment of Grant Number R-801874. The opinions, findings, and
conclusions expressed are those of the author and not necessarily those
of the Environmental Protection Agency.. Mention of company or product
names is not to be considered as an endorsement by the Environmental
Protection Agency.
Publication Number APTD-1498
11
-------
ABSTRACT
To facilitate controlled, reproducible testing for nitric-
oxide (NO) generation, a mechanical analog for the spark-igni-
tion was designed, instrumented and tested. The analog provides
for constant-volume combustion followed by product-gas ex-
pansion through a cam-driven poppet valve, giving pressure-
time histories like those of spark-ignition engines.
Measurements of NO concentrations in the product gas were
made several seconds after combustion both with and without
product-gas expansion. NO concentrations were independent
of the rate at which product gases were expanded. Ultraviolet
absorption by the (0,0) y band of NO (2265A) was used to'monitor
these NO concentrations.
Efforts to measure NO concentrations during constant-
volume combustion were impeded by unanticipated strong back-
ground absorption by C02. This rendered the combustion chamber
opaque during combustion and precluded monitoring of NO ab-
sorption. Absorption measurements at longer-wavelength absorp-
tion bands of NO (e.g., (0,2) y band at 2580 A) may escape such
"masking", but this possibility appears marginal, except for
drastically decreased absorption path lengths. Further efforts
in this direction were beyond the scope of the present, ab-
breviated program.
This report was submitted in fulfillment of Grant NO.
R-801874 under the sponsorship of the Office of Air and Water
Programs, Environmental Protection Agency.
111
-------
CONTENTS
Section Page
I Conclusions . 1
II Recommendations 2
III Preface 3
IV Introduction 4
V Analog Device 6
VI Nitric-Oxide Monitoring Technique 12
Post-Combustion Measurements 12
Measurements During Combustion 19
VII List of Publications 33
VIII References 34
IX Appendix 37
Program Listing . 50
y
-------
FIGURES
Page
Fig. 1 Automotive Engine Analog 7
Fig. 2 Poppet Value Detail 9
Fig. 3 Comparison of Spark-Ignition-Engine 11
and Analogue-Device Pressure Histories
Fig. 4 Steady-State Ultraviolet-Absorption 13
Apparatus
Fig. 5 Reference Gas Absorption Scans 11
Fig. 6 Nitric-Oxide-Absorption Calibration Curve 15
Fig. 7 Dependence of Citric-Oxide Product-Gas 16
Concentration on Initial Pressure
Fig. 3 Dependence of Nitric-Oxide Product-Gas 17
Concentration on Equivalence Ratio
Fig. 9 Flash-Lamp Circuit 20
Fig. 10 Spectral Distribution of Flash-Lamp Source 23
Fig. 11 Absorption by 4040 ppm-NO (in N2) 2H
Reference Gas (Total Pressure: 40 psig)
Fig. 12 Absorption by Products at Various Equivalence 25
Ratios (<1>) and Times Before and After Peak
Pressure (P )
max
Fig. 13 Absorption by Products of Stoichiometric 26
Combustion at Peak Pressure
Fig. 14 Absorption History During and After Firing 27
of Analog Device
Fig. 15 Absorption vs. Pathlength in Products of 32
Constant-Volume Combustion of Stoichiometric
Octane Air
VI
-------
Page
Fig. A-l Flow Chart for Equilibrium Constant- 41
Pressure Combustion Portion of Program
Fig. A-2 Flow Chart for Equilibrium Isentropic 42
Compression Portion of Program
Fig. A-3 Flow Chart for NO Kinetics Portion of 43
Program
Fig. A-4 Partial Sample Output from Program- 45
' Equilibrium and Kinetic Results After
Burning of 1st Shell
Fig. A-5 Partial Sample Output from Program- 46
Equilibrium and Kinetic Results After
Burning of 2nd Shell
Fig. A-6 Partial Sample Output from Program- . . 47
Kinetic Results Only, After Burning
of 10th (Last) Shell
Fig. A-7 Calculated Upper and Lower Limits of 48
NO Concentration
vn
-------
TABLES
Table 1 Estimates of Ultraviolet Absorption
at 2265A by NO,C02, 0 in Products
of Stoichiometric Constant^Volume
Combustion of CgH,g and CH.
Table 2 Program Listing
50
vi i-1-
-------
SECTION I
CONCLUSIONS
An approach for study of the mechanism of nitric oxide
(NO) production in spark ignition engines was developed
during the subject program. It involved the use of an ex-
perimental analog to the engine combustion and product gas
expansion processes and monitoring of NO concentration
during these processes by means of a spectroscopic technique
with fine-scale time resolution. The intention was to relate
NO generation and destruction to the engine processes in
order to estabish a rational basis for reducing its emission.
The simulator was developed, installed and operated
successfully. As expected, its operation was more repro-
ducible than that of an actual spark ignition engine, and
the range of parameter variation was broader than that of an
actual engine. It fills the gap between the more fundamental
research laboratory studies of NO kinetics (say, with a shock-
tube or well-stirred reactor) and the actual engine. It should
play a useful role in this regard.
The development of the time-resolved spectroscopic
technique was thwarted by a combustion product of unantici-
pated strength as a light absorber in the wavelength region
of critical interest. At the outset of the program there was
no indication in the technical literature that the interfer-
ing specie,excited-state C02, would play such a powerful role.
Following isolation of the origin of the difficulty with the
proposed technique, there was insufficient time for its ,
successful resolution prior to termination of the program.
-------
SECTION II
RECOMMENDATIONS ;
The subject experimental analog of the spark ignition
engine should be exploited for the production of data to
fill the gap between that from actual engines and that from
fundamental research laboratory apparatus, insofar as the
NO production and destruction mechanisms are concerned.
The subject NO-monitoring technique should be extended
to overcome the problems due to the presence of excited-
state CO- in the combustion problems.
The published work of others, for example, that of
Shahed and Newhall [27] which involves a similar NO-
monitoring technique and makes no mention of interference
by C0_, should be critically re-examined.
-------
V
SECTION III
PREFACE
This is a report of an investigation of nitric-oxide
(NO) generation in a simulated spark-ignition engine,
supported, in part, by Environmental. Protection Agency",'
Grant No. R-801874 (formerly PHS Research Grant No. 5R01
AP00847-D3).
The engine simulator design, performance and operating
characteristics has been reported in detail (Hodges, -J.L.,
Kurylko, L. and McAlevy, R.F., III, "Nitric Oxide Measure-
ment in a Simulated Spark Ignition Engine", Report No.
ME-RT-73008, Stevens Institute of Technology, Hoboken, N.J.,
January 31, 1971). It forms the framework for, and con-
tinuity with, the previously unreported material dealing
with computer calculations of NO in the simulator and NO
monitoring instrumentation. These activities were limited
by the .availability of funding.
-------
SECTION IV
INTRODUCTION
The automotive spark ignition engine is responsible for
a large fraction of the man-made nitric oxide found in the
lower atmosphere.- Nitric oxide at concentration levels
typical of urban environments catalyzes photochemical reactions
involving ultraviolet light and unburned hydrocarbons from
automotive exhaust gas. The results of these reactions is
the production of tnat conglomerate of chemical compounds
generally referred to as photochemical smog [1,2]. Nitric oxide
is also toxic and at typical urban levels may be detrimental
to nearby plant and animal life.
Early attempts to deduce the thermochemical processes
which lead to excessive nitric oxide emission from spark
ignition engines involved an analysis of the steady-state
exhaust-gas concentration of nitric oxide as a function of
such engine operating parameters as compression ratio, fuel-
air ratio, engine speed, etc. [3,4,5], These data, in con-
junction with kinetic schemes which evolved from furnace and
shock tube experiments [6,7], provided a basis for a kinetic
model of the formation and subsequent partial decomposition
of nitric oxide during the combustion and expansion of the
product gas in spark-ignition engines [8]. Steady-state
nitric-oxide measurement techniques have utilized the mass
spectrometer [3], spectrophotometer [4], and wet chemistry
[5].
There have been relatively few attempts to obtain time-
resolved NO-concentration measurements in spark-ignition
engines. A periodic sampling device was used to obtain quasi-
temporal measurements by Wimmer and McReynolds [8]and later
by Starkman [9]. Newhall related the absolute infrared gas
emission at the 5.3y fundamental of nitric oxide to NO-concen-
tration histories in spark-ignition engines [10]. LaVoie,
et al., have related NO-concentration histories in spark-ignition
engines to the continuous visible radiation resulting from a
photochemical reaction involving nitric oxide and atomic
oxygen [11]. Their experimental procedure was verified by
monitoring several wavelengths in the continuum and by adding
nitric oxide to the incoming charge. Extensions of these
techniques are described in References 19-21.
Most current theoretical studies of nitric-oxide formation
in spark-ignition engines assume that the Zeldovich mechanism
[6, 7], or a slight modification thereof (19-22), is applicable.
That is to say,
N2 + 0 sss- NO + N (1)
N + 02^s-NO + 0 (2)
N + OHssfeNO + H (3)
.J
-------
Generally, the oxygen atom is assumed to be in equilibrium
with its molecule and nitrogen atoms are assumed in the
steady state [11,19-22].
As a consequence of the strong temperature gradients
within the combustion chamber during and immediately after
combustion, there will also exist strong nitric-oxide-concen-
tration gradients. In a system of this complexity, in situ
experimental verification of the theoretical model is
essential. The spark ignition engine is, however, a poor
vehicle for these studies. Cycle-to-cycle variations in peak
cylinder pressure (up to 15 percent standard deviation [12])
can introduce an unacceptable degree of data scatter for any
concentration measurements. Many second-order effects, such
as spark advance, piston motion during combustion, and fall in .
volumetric efficiency with rpm, unnecessarily complicate
interpretation of experimental results. Additionally, the
flexibility of an engine is limited. The possible range of
variables, such as peak compression pressure, rate of ex-
pansion, and fuel-air ratio is low. For these reasons it was
decided to develop an experimental analog device which would
simulate, as closely as possible, pertinent phenomena occurring
in real engines, while eliminating irrelevant effects and being
capable of attaining a degree of precision not possible with
actual engines.
The analog device was also instrumented to allow spectro-
scopic measurements of NO concentrations within the chamber.
Ultraviolet absorption by NO at about 2260A was used first to
determine NO concentration in post-firing combustion products
in the chamber. Based.on these results, later spectroscopic
efforts were directed toward measuring NO-concentration
histories during combustion. ."-•'.
A computer program was developed in order to relate ex-
perimental observations from the analog device with a theoretical
description of the chemical-kinetics of NO formation. The
program was first developed to yield predictions of combustion-
chamber pressure and spatial distributions of temperature and
equilibrium species concentrations (including NO) (original pro-
gram, Ref. 23). Finally, modified Zeldovich kinetics for NO
formation were superimposed on previous equilibrium calculations
so as to allow prediction of local NO concentrations within
the chamber. Rather than including a model of flame-front pro-
pagation during combustion, the program relates thermodynamic
states within the combustion to time by way of input data for
the experimentally-observed pressure history in the chamber.
.5
-------
SECTION V
ANALOG OEVICE
In order to design an acceptable alternative, for
research purposes, to the spark-ignition engine/ it was
necessary to define first those disadvantages peculiar to
the spark ignition engine and then determine the basis for
those disadvantages.
There is considerable evidence that the principal
reason for excessive cycle-to-cycle variation in peak
engine pressure is the motion of the charge within the
combustion chamber prior to ignition [13,14]. That is, the
early rate of pressure rise appears a strong function of the
local gas velocity, in the vicinity of the spark plug, just
before ignition. Turbulence of the incoming charge will
result in more intimate mixing and an increase in flame-front
area, with an increase in flame velocity and a reduction
in susceptibility to detonation. However, pressure traces
obtained by igniting a quiescent fuel-air mixture should be
considerably more reproducible than those obtained from
engines; detonation of quiescent mixtures can be inhibited
by reducing flame path length. Additionally, an actual engine
is constrained to deliver at least enough useful work to
overcome frictional losses. It might be desirable, however,
to expand the combustion product gas more slowly than would
be possible with an engine. And thus expand the time scale
over which the NO kinetics can be observed.
The analog device/ then, is essentially a cylindrical,
constant-volume bomb (Fig. (1)). A spark plug is mounted in
one end of the combustion chamber, along the axis of symmetry,
with a cam-driven poppet valve in the other end. After com-
bustion the poppet valve is opened, allowing some of the
product gas to exit the chamber. By controlling the rate at
which the gas leaves the chamber, it is possible to modulate
cell pressure as a function of time. The product gas within
the combustion chamber does work to expel part of the gas
mixture rather than on a moving piston. The effect is, however,
analogous to the expansion of the product gas in an actual
engine.
The poppet valve and ignition breaker point cams are
mounted on a common shaft which is connected, through a solenoid-
operated wrap-spring clutch, to a continuously operating d-c
motor. The time interval between ignition and poppet valve
opening is controlled by adjusting the relative angular position
of the two cams. Valve lift was determined as a function of
the rate of change of gas pressure with time by coupling the
equations for choked flow and isentropic expansion, and the
perfect gas law. That is, assuming choked flow at the poppet
valve throat,
-------
spark plug
oir mlet
to vacuum pump
y
sapphire window
Combustion
Chamber
__..
1 — J
3
^
.
X
V
•MM
1 1
r ^\
©lectrie
motor
ajf-fcmjg
ignition cam
valve\
CQmJ
\
spring /
clutch^
from
control
circuit
Figure 1 Automotive Engine Analog
-------
then
T.
1/k
and
k/kRgT '
l (3k-l)/2k
o
k+1 \ 1/2
• / 2 ^ k~"^ I 1/7
w/A = \kg/R f k ^ . ) I P0/To
To/To1 = (Po/Po')(k~1)/k (5)
V = mRTn
/V P ' ^~-L^/^ \ (8)
dm/dt = L,i ° , ) (P0(1'k)/k)dPo/dt
V o /
combining equations (4^ and (8)
I/? /v
i '
P./T.
x (P0(1-k)/k)dP0
eliminating T0 and solving for A,
_ Vo cl(P0/P0 ')/dt
A (10)
The required flow area is determined from equation (10),
This is matched by the valve lift area, which is the surface
area of the conical frustum of slant height, t (see Fig. 2).
The valve lift area is related to the valve stem lift by
2 1/2
A = TTS sin a(d - sin a cos a) (1 + cos a) (11)
Combining equations (10)and (11):
s(d - s sin a cos a)
V0(d(P0/P0')/dt)/(P0/P0')
in a(l + cos2 a) 1/2(k) ARgT* (2/(k + jj ) U+D/2 (k-1)
-------
Figure 2 Poppet Valve Detail
-------
A computer program was developed to generate the required
cam profile for the valve lift requirements since conventional
graphical techniques were not sufficiently accurate.
Operation
To initiate a test cycle, a microsyringe was used to in-
troduce fuel to the evacuated combustion chamber. The chamber,
which was maintained at a constant temperature, was then pres-
surized with dry air. In this manner conditions were produced
which corresponded to those in an engine just prior to ignition.
Engagement of the clutch resulted in precisely one revolution
of the cam shaft. As the shaft rotated, the ignition breaker
points closed, igniting the fuel-air mixture.
Performance
A comparison of typical pressure histories obtained with
the analog device and an actual CFR spark ignition engine is
presented in Fig. (3). The analog device pressure traces are
considerably more reproducible than the engine pressure traces.
The rates at which pressure rises and falls during combustion
and expansion are slightly less with the analog device than with
the engine. This could have an influence on the kinetics of
nitric oxide formation and decomposition during combustion [22].
The lower rate of pressure rise in the analog device is a con-
sequence of the low flame speed typical of quiescent gas mix-
tures. The maximum rate of expansion is limited by the mechanical
design of the apparatus, the practical limit being imposed by
maximum inertial forces in the valve-cam system.
Pressure traces were obtained at initial pressures between
25 and 125 psia, and initial temperatures between 250 and 450
deg. F. Most data were obtained with the initial temperature
of 350 deg. F. Peak measured pressures varied between approxi-
mately 125 and 650 psia. In the analog the volumetric efficiency
was 100 percent and the combustion proceeded at constant volume.
This resulted in peak pressures generally higher than those en-
countered in an engine where residual gas causes dilution of
the active charge and where piston ring blow-by would offset
maximum pressures.
Iso-octane fuel was used in all runs.
No detonation was observed in any of the tests.
10
-------
8:1 COMPRESSION RATIO
980 RPM
100 psi/div
5 ms/div
INITIAL TEMPERATURE: 350 F
INITIAL PRESSURE: 125 psia
EQUIVALENCE RATIO: 1.1
100 psi/div
20 ms/div
Note: Analog device pressure traces represent superposition
of constant volume and expansion runs
Figure 3 Comparison of Spark-Ignition-Engine
and Analog-Device Pressure Histories
i'
11
-------
SECTION VI
NITRIC-OXIDE MONITORING TECHNIQUE
Post-Combustion Measurements
A new monitoring technique was developed to measure nitric
oxide product gas concentration a few seconds after combustion.
The technique was used in place of established methods because
it was hoped that it could be eventually extended to obtain
time-resolved in situ measurements during combustion.
Nitric-oxide concentration in the product gas was related
to the absorption of ultraviolet light at the (0,0)y band of
nitric oxide between 2260 A and 2270 A [15]. A schematic of
the monitoring apparatus is presented in Fig. (4). It was found
that, at a fixed temperature, the absorbance of the gas mixture
was a function only of the partial pressure of nitric oxide.
At low NO partial pressures the measured absorption coefficient
agreed with published values [16].
Absolute or integrated band intensities were calculated
and compared with published data [17]. Use of a single beam
monochromator made acquisition of absolute intensities difficult
and imprecise. Allowing for this, the data correlated reasonably
well with the literature.
As an experimental control, calibration curves were obtained,
using a reference gas (a mixture of 4040 PPM NO, balance N,),
prior to and after each series of experimental runs. Reference
gas absorption traces are presented in Fig. (5). The calibration
curves consisted of plots of the reference gas absorbance be-
tween and at each of the two band heads versus the known nitric
oxide partial pressure (Fig. (6)). Subsequently, the absorbance
of the combustion product gas at the two band heads was measured,
and, using the calibration curves, the corresponding partial
pressure of nitric oxide was determined. Measurements of the
total pressure of the product gas following combustion permitted
the calculation of the nitric oxide concentration in the exhaust
gas as a function of such parameters as initial pressure, initial
temperature, equivalence ratio and rate of expansion.
Discussion
As expected, NO concentration was a weak function of
initial temperature and pressure (Fig. (7)), and a strong func-
tion of equivalence ratio (Fig. (8)). Evident is the "inverted
U" characteristic of measurements made on actual spark igni-
tion engines [3,4,8]. Measured NO concentrations were some-
what higher than would be expected in engines. There are numer-
ous possible explanations for this. First, the residual exhaust
gas fraction carried over from one cycle to the next will tend
to dilute the incoming charge, with a consequent reduction in
12
-------
XENON
ARC
COMBUSTION
CHAMBER
QUART! LENS
^^•^^H
7
RECORDER
SPECTROMETER
PHOTO-
MULTIPLIER
Figure, 4 Steady-State Ultraviolet-Absorption Apparatus
-------
Figure 5 Reference-Gas-Absorption Scans
-------
1.0
O.I 0.2 0.3 0.4 0.5 0.6
PARTIAL PRESSURE OF NITRIC OXIDE (PSD
0.7
Figure 6 Ni'tric-Oxido-Absorpticri Calibration Curve
-------
I
10,000
8000
6000
50
!
U, 4000
I
2000'
o
o
EQUIVALENCE R/
INITIAL TEMPER;
o CONSTANT VOLU
=»THEORE
EQUILIE
LIMITS
STAND/1
S
TICAL PEAK TEMPERA
JRIUM
REPRESENT TWO
^RD DEVIATIONS
\TIO= 1,0.
&TURE' 350° F
ME COMBUSTION
I
TURE
2
20
40 60 80 100
INITIAL GAS PRESSURE (PSIA)
120
140
Figure 7 Dependence of Nitric-Oxide Product-Gas
Concentration on Initial Pressure
-------
I
I
1
!
»U
1
|
>,000
o
0
o
o
CD
"0
o
o
a>
"0
o
o
A
0
o
o
TO
0
o
o
THEORETICAL PEAK TEMPERATURE
EQUILIBRIUM
LIMITS REPRESENT TWO
STANDARD DEVIATIONS
a
INITIAL TEMPERATURE'350°F
INITIAL PRESSURE = 75 PSIA
o CONSTANT VOLUME COMBUSTION
® COMBUSTION FOLLOWED BY EXPANSION
.8 .9 1.0
EQUIVALENCE RATIO
I.I
1.2.
1.3
Figure 8 Dependence of Nitric-Oxide Product-Gas
Concentration on Equivalence .Ratio ..
-------
peak cycle temperature (this effect is consciously used to
reduce nitric oxide emissions from engines in the familiar ex-
haust gas recycle technique). In the analog device, the
residual exhaust fraction is virtually zero. The presence of
moisture in the air similarly suppresses peak cycle tempera-
tures [18]. Dry air was used in all experimental runs described
herein. Additionally, in engines piston ring blow-by and piston
motion allow the gas to expand during combustion, again tending
to reduce peak cycle temperature. These effects are also ab-
sent in the analog device, and so, as a consequence of the in-
crease in peak cycle temperature, higher measured nitric oxide
concentrations would be expected.
A particularly interesting effect was noted when measured
nitric oxide concentrations obtained when the product gas was
expanded as rapidly as possible following combustion were com-
pared with nitric oxide concentrations in the product gas after
constant volume combustion. The measured nitric oxide concen-
trations were virtually the same in each case. This would tend
to imply that piston motion plays a relatively minor role in
the events which lead to the freezing of nitric oxide in spark
ignition engines.
Current kinetic schemes invariably assume that the only
means by which a reduction in product gas temperature can be
obtained is the isentropic expansion of that gas (i.e., through
piston motion). Effects such as heat transfer to the wall,
radiation, etc. are thought, quite correctly, to have only a
minor influence on the bulk temperature of the gas. However,
it should be noted that strong temperature gradients (up to
400 deg F) exist across the combustion chamber. Nitric oxide
equilibrium concentration in the hottest part of the product
gas is about six times that in the coolest part [11]. That por-
tion of the gas mixture richest in NO is closest to the wall,
where wall cooling effects would be expected to have the greatest.
influence. Additionally, any mixing of the hot and cool seg-
ments of the product gas could also result in freezing of the
nitric oxide. Apparently the "rate of expansion", or more
specifically, the rate at which gas pressure changes with time,
is reflective only of the rate at which the average, or bulk,
temperature changes with time. Local cooling rates, however,
would be expected to exceed the average value in precisely
that portion of the mixture richest in nitric oxide.
18
-------
Measurements During Combustion
The approach described above and in Ref. 23 was adapted
to measuring NO concentrations in_ situ during combustion.
This direction was indicated by~The results of post-combustion
NO measurements [23]. These included the observation that NO
"freezing" in product gas did .not appear .to depend strongly on
piston motion but rather on processes occurring during com-
bustion.
Preliminary spectrographic data were taken during com-
bustion by replacing the steady arc-lamp and the photomulti-
plier used in post-combustion measurements (see Fig. 4) with
a Xenon flash-tube and spectrographic film, respectively.
The experiments.involved obtaining a spectrograph by repeated
exposures to the flash-lamp source. In order to achieve
suitably high densities on the spectrographic film, the light
source was fired (repetitively) at selective times either
during or after combustion for a succession of replicate
firings of the analog device. These preliminary measurements,
reported in detail in Ref. 23, were not sensitive enough to
discern NO prior.to 5 msec before the attainment of peak
chamber pressure (i.e., 5 msec before the completion of com- .
bustion).
Apparatus
, To adopt the previously used equipment to NO measurement
during combustion, substantial improvements in auxiliary equip-
ment (exhaust hoods, etc.) and in mounting and positioning
equipment (for. optics, analog device, etc.) were made. These
design changes were dictated by both safety and optical align-
ment .considerations , given the decision to continue usage of
the analog device and spectrometric instrumentation over an
extended time period. The major apparatus development re-
quired, however, was of the instrumentation for time-resolved
NO measurements.during combustion.
The spectrometric instrumentation developed included a
new flash-lamp source, modification of the associated trigger
circuit, and a new detection system. These new developments
were dictated by the desire for a higher sensitivity system
which did not require multiple firings of the analog device
for each data point; the multiple firings required,in pre-
liminary spectrographic measurements (see above) were judged
to be impractical.
The flash-lamp source which was developed for time-resolved
NO measurements is shown schematically in Fig. 9. The power
source for charging the capacitor was an Ascorlight Model
62-509 high-voltage power supply which usually was operated at
about 1000 vdc yielding.flash energy inputs of about 450 watt-
seconds. Linear-pulsed xenon flash tubes (EG & G Models
FX42C-3 and FX38A-3) provided a 3" are of roughly 1000
19
-------
V\AA
1.2 n
40W
A
EGSG
FX 38-3
OR
FX 42C-3
FLASH TUBE
FROM TRIGGER
MODULE
Figure 9 Flash-Lamp Circuit
-------
horizontal-candlepower seconds. The arc was focussed on the
spectrometer entrance slit following collimation by a quartz
lens and transmission through the sapphire windows of the
analog device. The optics, flash-triggering, and time-delay
circuit were essentially those detailed in Ref. 23 excepting
minor modification of the .time-delay circuit. Alignment of
the optical system was facilitated by a back light, a 12 vdc ,
coiled tungsten filament lamp built into the flash-tube
mounting. This allowed imaging of the flash tube on the
entrance slit of the spectrometer .during alignment. The
flashes used during NO measurement were of approximately 0.5
msec duration (between 50%-intensity points) and were of ample
energy to be detected by the spectrometer detector.
The previously used spectrometer (Jarrell-Ash Model
82-020, 0.5 meter, Ebert type) was equipped with a grating of
1200 lines/mm (3000A blaze) and entrance.and exit slits of 50
yn, yielding a spectral slit width of about 1.5A(Ref. 23, p.
118). The f-number of the spectrometer (f/8.6), rather than
the associated optics, limited transmission. Tests indicated
a hysteresis(backlash) of approximately 1A in the wavelength
setting of the monochromator.
The previously used photomultiplier detector [23] was.
found to be excessively noisy presumably owing to age. and
usage. A new photomultiplier (EMI Model 9558Q) with fused-
silica window., a new photomultiplier housing .(EMI Model E258)
and a potted dynode chain (EMI Model D113 FL) were adapted to
the spectrometer exit aperture. The photomultiplier was
powered by a Brandenburg Model 500 power supply, typically
operating at 800 to 900 vdc. Photomultiplier output was re-
corded photographically using an oscilloscope (Tektronix
Model 564, with Type 3A3 Dual-Trace Differential Amplifier and
Type 2B67 Time Base). Linearity of response of the detection
system was verified using calibrated neutral-density filters
(quartz substrate).
21
-------
Discussion
Preliminary testing with the new instrumentation and pro-
cedures for time-resolved NO measurements revealed several
problems with reproducibility of flash-lamp-spurce peak intensity
and triggering. These problems were sidestepped temporarily
with the aim of proceeding as quickly as practical to further
evaluation of the proposed approach to NO measurement. Re-
producibility of flash-lamp intensity and triggering, while .not
completely satisfactory, were viewed as good enough to allow
rough measurements of NO concentrations during firing of the
analog device.
Figure 10 shows the results of successive flash-lamp firings
with the test chamber of the analog device evacuated. The spectral
distribution and scatter shown are representative of the flash-
lamp system used in further testing. While reproducibility is
not totally satisfactory, the flatness of the spectral distribu-
tion of the lamp radiation is helpful.
Figure 11 shows the spectral distribution of photomultiplier
output when the combustion chamber of the analog device was
pressurized to 40 psig with a reference gas of 4040 p.p.m. NO -
balance N2. Absorption of radiation from the flash lamp by the
two-headed (0,0) y band of NO centered at about 2265A is evident.
Despite some shortcomings in reproducibility of flash-lamp
firing, a series of test firings of the analogue device was
made to allow better evaluation of the potential of the time-
resolved ultraviolet-absorption approach for measuring NO con-
centration.
Test firings of the analog device were made at various
equivalence ratios, pre-firing pressure and temperature of 4.4
atm and 450 K. Measurements of ultraviolet absorption by the
products of combustion were made at various times during com-
bustion (before attainment of peak pressure) and after. In
each case comparable measurements were made of photomultiplier
output with the test chamber evacuated.
Figures 12 and 13 show the results of spectral-absorption
measurements at various times and equivalence ratios. Noteworthy
is the apparently high absorptivity of the combustion products
over a wide spectral range. This had not been observed in earlier
measurements on products well after completion of combustion
when the products are cooler and at lower pressure due to heat
loss (Ref. 23, also Fig.14 ). Some effect of this type had
been expected owing to the presence of 02- However, it was
expected that this absorption would not substantially exceed
that of NO and that, outside the (0,0) Y band of NO the absorp-
tivity of the products would not exceed approximately 50% [23].
22
-------
NJ
U)
>
SE
1- 4
Q.
§ 3
cc
UJ
_J
a.
5 2
ID
2'
O
O i
I '
Q.
n
•
A,^
EVACL
JATED
TEST CHAMBER
. •• ' .
» «
«
'
f T
I. . I
T 1 RANGE OF DATA (
X
I.
4
" I
--I-- ••*
X
3 TO 5 TESTS)
T
' ^^
•
2220
2240 2260
WAVE LENGTH (A)
2280
Figure 10 Spectral .Distribution of Flash-Lamp Source
-------
KJ _'
r
PHOTOMULTIPLIER OUTPUT (ARB. UNITS)
~ P P p p -
O M .&> fft CO O
—
—
—
—
AT-
1 A ' ' '
a
&
0 (0,
o
1 1 1 1
1 1 I.I
<
0)7 BAND L
SO
o
'S«a ^
9 ii i
i i i
**
>
A
1 1 1
WAVE LENGTH (A)
Figure 11 Absorption by 4040 ppm-NO (in N2)
Reference Gas - Total Pressure: 40 psig
-------
NJ
tn
CD
CC
O
=I., 15 MSEC BEFORE PMAX
D =0.9, 9MSEC BEFORE PMAX
1 1
2250
2270
WAVE LENGTH (A)
2280
Figure 12 Absorption by Products at Various Equivalence Ratios
and Times Before and After Peak Pressure (
-------
% ABSORPTION
W *. O GO C
5 O O O O C
-=» . —
—
—
—
MM
200
2300 2400
WAVE LENGTH (A)
2500
Figure 13 Absorption by Products of Stoichiometric Combustion
at Peak Pressure
-------
IUU
80
S 60
O
CD
4j
20
0
t
»
o
I
—
—
—
\
\
1 1 II
till
\
\
%
1 1 1 1
1 1 1 1
\^
^^«
1 1 1 1
1 1 1
— •
—
—
^"-o- ~
1 1 lA 1
0 100 200 300 V5C
> -
>00
TIME AFTER PEAK PRESSURE (M SEC)
Figure 14 Absorption History During and After Firing of Analog Device
-------
Since the concentration of 0~ in combustion products is
strongly dependent on equivalence ratio, absorption due to
the Schumann-Runge bands of 0~ should decrease appreciably as
the equivalence ratio is increased toward richer combustible
mixtures. Tests of rich mixtures in the analog device did
not, however, evidence substantial decreases in ultraviolet
absorption outside of the (0,0) y band of NO. Hence, it ap-
peared that either or both of two effects were probably re-
sponsible for the broad-band ultraviolet absorption observed
in test firings: (i) disruption of the beam through the test
section, perhaps due to refraction by flame-front-generated
turbulence, or (ii) presence of an unknown broad-band absorbing
species.
The possibility of massive disruption of the spectrometric
light beam by refractive effects was tested by passing the nar-
row beam of a CW helium-neon laser through the test cell and
onto a small photodetector (ca. 2 mm diameter). The photo-
detector output was recorded during a test firing, and no
major change in signal level was observed. This indicated that
beam displacement after passage of the flame front was negli-
gible and not a likely cause of the observed broad-band de-
creases in photomultiplier output during NO-measurement tests.
A search for possible broad-band absorbers other than 0,,
ensued. Carbon-or hydrogen-containing species were focussed
on owing to the previously mentioned fact that the observed
broad-band ultraviolet absorption was found to be relatively
independent of equivalence ratio. Standard references on
absorbing/emitting species in flames proved fruitless. Finally,
late in the second year of this program, a little-recognized
broad-band ultraviolet absorber was identified: CO?'
Broad-Band Interference (Ultraviolet Absorption) by CO2
Recent shock-tube studies of the chemical kinetics of NO
formation [24] uncovered the fact that the CO- molecule ex-
hibits broad-band ultraviolet absorption at elevated tempera-
tures [25]. Apparently this absorption has not been widely
cited in the combustion literature because of the paucity of
spectrometric combustion studies at the short ultraviolet wave-
lengths of concern in the present investigation.
Based on the available data for the ultraviolet absorption
of C02, estimates were made of the ultraviolet (2265A) absorp-
tivity of the products of constant-volume combustion of stoichi-
ometric octane-air mixtures. The optical path lengths considered
were those corresponding to the analog device (physical path
length: 8.25 cm.). Similar calculations were made for a fuel
of lower C/H ratio, CH4. The results are shown in Table 1.
The results in Table 1 show that the high absorption coef-
ficient of the (0,0) y band of NO, as measured earlier in the
present study [23], render the combustion products in the analog
1 V ' ' •
28
-------
Fuel
(with
air)
C8H18
"
"
CH4
11
11
Species
NO
co2
°2
NO
co2
°2
k,
Absorption
Coefficient
(cm )
30{a)
7{b)
4.0/C>
1.2(d)
30 (a)
-, (b)
/
4.0(c)
1.2{d)
x,
Mole
Fraction
.022
.072
.022
.018
It
.057
.018
Optical
PathLength
(cm)
0.60
n
1..95
0.60
.45
II
1.43
.46
a ,
Absorption
99+
92
99+
51..
99+
85
99 +
42
Notes: (a) Ref. 23- •.
(b) Ref. 16 a = 1 - exp (- k JL ' x ^2)
(c) Ref. 25 r = | § x <8-25 cmV
std std
N
(d) Ref. 23 j— = 0.61 (depopulation of ground state of NO)
TABLE 1 Estimates of ^Ultraviolet Absorption at 2265A by
Individual Species in Products of Stoichiometric
Constant- Volume Combustion of
, „ and CH.
29
-------
device optically thick for narrow spectral slit widths centered
in the (0,0) y band. However, in the absence of major "back-
ground" absorption by C0_ or 0~, this does not render NO con-
centration measurements unfeasible. As was demonstrated
earlier in the present program (Ref. 23, p. 76) large spectral
slit widths, exceeding the NO-band width, render the test gas
transparent enough to allow absorption measurements and, hence,
NO-concentration measurements.
Unfortunately, however, the results of Table 1 show very
high absorption by C02 as a "background" absorber. This absorp-
tion is over a much broader spectral range than that of the NO
band (Ref. 25, also Figure 12). Thus, in contrast with NO
absorption, C02 absorption is not substantially decreased by
increasing the spectral slit width for absorption measurements,
i.e., C02 absorption, being less spectrally localized than NO,
renders the test gas opaque even for large spectral slit widths
centered at the (0,0) y band of NO. This apparently precludes
absorption spectrometry in C02~containing product gas in the
neighborhood of 2265A.
Only two alternatives are apparently possible to escape
the obstacle of CO2 absorption: (i) moving to a different NO
absorption band where C02 absorption is much lower, rendering
the products optically tnin, or (ii) decreasing the physical
(and, therefore, optical) path length in the analog device.
Relative to the possibility of NO absorption spectrometry
at higher wavelengths, it is noteworthy that CO- absorption
apparently extends little diminished (see Fig. 13) to 2500
to 2600A. Calculations like those for Table I were made using
the absorption coefficients of Bowman and Seery [24] at 2358A
for CO2. The results showed only slight improvement («ro =0.98)
at this wavelength, corresponding with the (0,1) y band ^
of NO. The wavelength of the (0,2) y (2580A) may show some
further decrease in C02 absorption but probably only marginally
so considering the slow decrease in the absorption coefficient
of C02 with increasing wavelength [25]. Above this wavelength
other interfering species, e.g., OH and CO [26] begin to be in-
fluential so that little, if any, "window" exists, i.e., com-
bustion products (other than NO) in the analog device are likely
to be optically thick and absorption spectrometry is not practical.
Detailed consideration of the spectrometric potential of the
(0,2) or higher wavelength y bands for NO concentration measure-
ment might show marginal value, but further investigation, ex-
perimentally or theoretically, was precluded by truncation of the
present investigation.
The possibility of decreasing the optical thickness (path
length) in combustion products of the analog device is also not
promising as a route for minimizing interference by the strongly
absorbing C02 molecule. The physical path lengths required to
30
-------
allow transmission of a substantial fraction of the spectro-
metric beam are too short to be practical for a device in-
tended to analogous to a spark-ignition combustion chamber;
physical path lengths on the order of less than 2 cm. would
be required (Fig. 15).
In summary, then,it appears based on both theoretical
and experimental data that ultraviolet absorption by CO- pre-
cludes spectrometric measurement using the y bands of NO to
determine NO concentrations in high-temperature combustion
products of hydrocarbon fuels burned in practical-sized com-
bustion chambers. This conclusion contradicts the basis for
the measurements reported by Newhall and Shahed [27] but ap-
pears to be firmly grounded. The failure of these earlier
workers to observe or consider C02 absorption is currently
unexplained. The above conclusion, however, does not preclude
the possibility of spectrometric NO measurements using the
Y bands of NO for non-hydrocarbon (e.g., H.-,) combustion sys-
tems. Of course, optically-thin configurations with dilute,
non-combustible mixtures such as those of the shock-tube in-
vestigations of Bowman and Seery [24] are also practical appli-
cations of ultraviolet, spectrometric NO detection and measure-
ment .
31
-------
100
OJ
KJ
PRESENT
ANALOG
DEVICE
3456
PHYSICAL PATH LENGTH (CM )
Figure 15 Absorption vs. Pathlength in Products of Constant-Volume
Combustion of Stoichiometric Octant/Air
-------
SECTION VII
LIST OF PUBLICATIONS
Hodges, J.L., McAlevy, R.F., III, Potter, J.H., and Kurylko, L.,
"Nitrogen-Oxide Kinetics in Auto Combustion Products - An
Interim Progress Report", Report No. ME-RT-70003, Dept. of
Mech. Eng., Stevens Inst. of Tech., Hoboken, N. J., July.1970.
Hodges, J.L., McAlevy, R.F., III, and Potter, J.H., "Nitric-
Oxide Generation in a Simulated Spark-Ignition Engine",
J. for Eng'g. and Industry (A.S.M.E. Transactions, Series B) ,
Vol. 93, pp. 432-436, 1971. -
Hodges, J.L., Kurylko, L., and McAlevy, R.F., III, "Nitric-Oxide
Measurements in a Simulated Spark-Ignition Engine", Report
No. ME-RT-70008, Dept. of Mech. Eng., Stevens Inst. of Tech.,
Hoboken, N.J., January 1971.
33
-------
SECTION VIII
REFERENCES
1. Bufalini, J., and Stephens, E., "The Thermal Oxidation of
Nitric Oxide in the Presence of Ultraviolet Light",
International Journal of Air and Water Pollution, Vol. 9,
1965.
2. Eschenroeder, A., and Martinez, J., "Mathematical Modeling
of Photochemical Smog", AIAA Paper, No. 70-116, 1970.
3. Campau, R., and Neerman, J., "Continuous Mass Spectrometic
Determination of Nitric Oxide in Automotive Exhaust", SAE
Paper No. 660116, 1966.
4. Nicksic, S., and Harkins, J., "Spectrophotometric Determina-
tion of Nitric Oxide in Auto Exhaust", Analytical Chemistry,
Vol. 34, No. 8, 1962.
5. "Selected Methods for the Measurement of Air Pollutants",
U.S. Department of Health, Education, and Welfare PB 169677,
1965.
6. Kaufman, F., and Kelso, J., "Reactions of Atomic Oxygen and
Atomic Nitrogen With Oxides of Nitrogen", Seventh Symposium
on Combustion, 1959.
7. Wray, K.L., and Teare, J.D., "Shock Tube Study of the Kinetics
of Nitric Oxide at High Temperatures", Journal of Chemical
Physics, Vol. 36, No. 10, 1962.
8. Wimmer, D., and McReynolds, L., "Nitrogen Oxides and Engine
Combustion", SAE Transactions, Vol. 70.
9. Starkman, E., Stewart, H., and Zvonow, V., "An Investigation
into the Formation of Emission Precursors", SAE Paper No.
690020, 1969.
10. Newhall, H., "Theoretical and Experimental Investigation of
Chemical Kinetics During Rapid Expansion of High Temperature
Combustion Products", PhD dissertation, University of Califor-
nia, Berkeley, 1966.
11. LaVoie, F., Heywood, J., and Keck, J., "Experimental and
Theoretical Study of Nitric Oxide Formation in Internal
Combustion Engines", Combustion Science and Technology,
Vol. 1, 1970.
12. Patterson, D., "Cylinder Pressure Variations, A Fundamental
Combustion Problem", SAE Paper No. 660129, 1966.
13. Cole, D., and Mirsky, W., "Mixture Motion - Its Effect On
Pressure Rise In a Combustion Bomb: A New Look at Cyclic
Variation", SAE Paper No. 680766, 1968.
7 > -i
34
-------
14. Peters, B. , and Borman, G., "Cyclic Variations and Average
Burning Rates in a S.I. Engine", SAE Paper No. 700064, 1970.
15. Herzberg,,G., Spectra of Diatomic Molecules, Van Nostrand
Co., 2nd ed., 1963.
16. Sullivan, J. , and Holland, A., "A Congeries of Absorption
Cross Sections for Wavelengths Less 'Than 3000 X", NASA CR-371,
1966.
17. Penner, S., Quantitative Molecular Spectroscopy and Gas
Emissivities, Addison-Wesley Publishing Co., Boston, Mass.,
1959.
18. Brown, W., Gendernalik, S., Kerley, R., and Marsee, F. ,
"Effect of Engine Intake - Air Moisture on Exhaust Emissions",
SAE Paper No. 700107, 1970.
19. Spadaccini, L.J., and Chinitz, W., "An Investigation of
Nonequilibrium Effects in an Internal Combustion Engine",
Journal of Engineering for Power (Transactions of the ASME),
April 1972, pp. 98-107.
20. Heywood, J.B., Mathews, S.M., and Owens, B., "Predictions
of Nitric Oxide Concentrations in a Spark-Ignition Engine
Compared with Exhaust Measurements", Paper No. 710011,
presented at the ASE Automotive Engineering Conference,
Detroit, Mich., Jan. 11-15, 1971.
21. Cavetto, L.S., Muzio, L.R., Sawyer, R.F., and Starkman, F.S.,
"The Role of Kinetics in Engine Emission of Nitric Oxide",
Combustion Science and Technology, 1971, Vol. 3, pp. 53-61.
22. Muzio, L.J., "Theoretical and Experimental Investigation of
Temperature and Composition During Combustion in a Closed
System", Doctoral Dissertation, University of California,
Berkeley, 1971,(University Microfilm #71-15 ,849).
23. Hodges, J.L., Kurylko, L., and McAlevy, R.F. Ill, "Nitric
Oxide Measurement in a Simulated Spark Ignition Engine",
Report No. ME-RT-73008, Stevens Institute of Technology,
Hoboken, N.J., January 31, 1971.
24. Bowman, C.T. and Seery, D.J., "Investigation of NO Forma-
tion Kinetics in Combustion Processes: The Methane-Oxygen-
Nitrogen Reaction" , Paper presented at Symposium on Emis-
sions from Continuous Combustion Systems, General Motors
Research Labs., Sept. 27-28, 1971.
25. Generalov, N;A. , Losev, S.A., and Maksimenko, V.A., "Ab-
sorption of Ultraviolet Radiation By Strongly Heated Carbon
Dioxide Gas", Optics and Spectroscopy, Vol. 15, pp. 12-14,
1963.
35
-------
26. Lavoie, G.A. , Heywood,. J.B., and Keck, J.C., "Experimental
and Theoretical Study of Nitric Oxide Formation in Internal
Combustion Engines", Comb. Sci. and Tech., Vol. 1, pp. 313-
326, 1970.
27. Shahed, S.M. and Newhall, H.K., "Kinetics of Nitric Oxide
Formation in Propane-Air and Hydrogen Air-Diluent Flames",
Comb, and Flame, Vol. 17, pp. 131-137, 1971.
28. JANAF Thermochemical Tables, Publication PB 168370, National
Bureau of Standards, August 1965.
29. Zeleznik, F.J. and Gordon, S., "A General IBM 704 or 7090
Computer Program for Computation of Chemical Equilibrium
Compositions, Rocket Performance, and Chapman-Jouguet
Detonations", NASA Technical Note D-1454, October 1962.
30. Bowman, C.T., "Investigation of Nitric Oxide Formation Kinetics
In Combustion Processes: The Hydrogen-Oxygen^Nitrogen
Reaction", Comb. Sci. and Tech., Vol. 3, pp. 37-45, 1971.
36
-------
SECTION IX
APPENDIX
37
-------
COMPUTER PROGRAM FOR
NITRIC-OXIDE CONCENTRATION CALCULATIONS
Purpose
A FORTRAN IV computer program was written for several
purposes:
(1) In its original form [23] , the program dealt only
with calculations of equilibrium NO concentrations
in a constant-volume combustion chamber. The aim
of these calculations was to allow comparison with
NO concentrations measured post-combustion in the
analog device.
(2) The original equilibrium program has been modified
slightly for greater precision and flexibility in
use and to include kinetic calculation of NO con-
centrations based on(partial) equilibrium con-
centrations of other combustion products. The aim
of this extension is to allow comparison of theo-
retical and experimental NO concentrations in the
analog device. The basic calculations performed
are those for NO concentrations during combustion
of hydrocarbon fuels under "normal" conditions
representative of automobile-engine combustion
after the compression stroke.
(3) As the program was being developed, it appeared
that extension of the program to additional
capabilities would be desirable. For example, non-
hydrocarbon fuels, reactants representing recircu-
lated combustion products, etc. are of current
interest. This capability would have allowed
broader application of the spark-ignition engine
analog device. The abbreviated project scope did
not ultimately allow such extension.
Summary of the Program
The program described briefly in this Appendix has
the capability for calculating equilibrium product-gas
compositions, pressures, and temperatures during and follow-
ing shell-wise, constant-volume combustion of hydrocarbon
fuels from arbitrary initial conditions of equivalence
ratio, pressure, and temperature. The chemical species
considered are limited to NO, N2, N, 02, 0, H2/ H, HjO, OH,
CO, and C0-
38
-------
In addition to equilibrium calculations and based upon
them, kinetic calculations of NO formation are included in
the program. In these calculations, a "partial-equilibrium"
approach is taken; all species excepting NO are considered
to be at equilibrium and NO kinetics are treated via the
steady-state approximation for the N atom. Modified
Zeldovitch kinetics are considered (Equations (1), (2), and
(3); page 7). The OH + N reaction is considered only for
the fuel-rich systems.
Provision is also made in the program for outputing the
thermodynamic state achieved after homogeneous constant-
pressure or constant-volume combustion (of a single "shell")
of reactants. This capability was included for its utility
in debugging and checking results against published values
and with data generated by other programs for more compre-
hensive thermochemical calculations.
The time scale for kinetic calculations is established
via empirical data, specifically via experimental pressure-
time histories from the analog device. The program stores
and interpolates data for the pressure-time history. The
calculated kinetics of NO formation and corresponding
equilibrium values for non-NO constituents are linked via
the pressure-time history. NO concentrations are calculated
at those times when the chamber pressure equals the various
chamber pressures calculated to exist after combustion of
each individual shell of reactants.
Since the combustion and NO formation are considered to
occur stepwise (shell-wise), the equilibrium states of the
various shells comprising the total volume and the NO con-
centration in each are calculated only at discrete times,
i.e., at those times corresponding to the end of burning of
each shell. At each such time, two NO concentrations are
calculated:
(i) the NO concentration calculated (kinetically)
assuming NO formation began at the beginning
of burning of each shell (overestimate, upper
limit)
(ii) the NO concentration calculated (kinetically)
assuming NO formation began only upon
completion of burning of each shell (under-
estimate, lower limit).
39
-------
In each of these cases, the temperatures, pressures and
concentrations of non-NO species which are used for
calculating the NO concentration are those existing at the
end of time period over which NO formation is being calcu-
lated.
Figures A-l, A-2, and A-3 diagram the logic of the
program. From the figures, it is evident that the equi-
librium combustion is treated as a shell-wise constant-
pressure combustion followed by isentropic compression of
both burned and unburned shells to maintain total chamber
volume constant.
Program Inputs
Input values are supplied entirely via FORTRAN IV
"DATA" statements in the program (see "Program Listing",
below). The following quantities are input:
FORTRAN
SYMBOL
PHI
P
TI
GARB
HYD
NS
NPT
81,82,33
C1,C2,C3
MM
Equivalence Ratio
Initial pressure of reactants
" temperature "
Carbon atoms per fuel molecule
Hydrogen "
No. of shells into which chamber is
subdivided
Coefficients for calculating fuel 2
enthalpy from H = Bl + B2 . T + B3 . T
Coefficients for calculating fuel entropy
from S = Cl + C2 • Jin T + C3 • T - R Jin P
Code no
"0" =
111 " =
H T H =
II O II _
, for mode of calculation
only equilibrium results are output
only NO kinetics results are output
both equilibrium and kinetic results
are output
only constant-pressure and constant-
volume results for burning of single
shell are output
PC
TE
Chamber pressures (in atm.) from empirical
pressure-time history from analog device
scaled to make peak pressure agree with
calculated peak pressure.
Times corresponding to chamber pressures
from analog device
-------
SET INITIAL
CONDITIONS IN ALL ,
THE SHELLS
I
CALCULATE THE ENTHALPIES H
OF THE REACTAKTS OF. THE
Subroutine ENTH
MAKE INITIAL ESTIMATION OF !
THE TEMPERATURE £ C0p, R00
CONCENTRATION OF eL
THE BURNING SHELL I
CALCULATE THE CONCENTRATION OF. i
THE'PRODUCTS OF THE BURNING SHELL!
I AT THE ESTIMATED TEMPERATURE ;
Subroutine CONG
CALCULATE THE ENTHALPIES H j
OF THE PRODUCTS OF THE ?j
BURNING SHELL I . I
ADJUST THE
TEMPEPJITURE OF
THE BURNING
SHELL I BY
ENTHALPY BALANCE
Subroutine iiNTH
Figure A-l Flow Chart for Equilibrium, Constant-Pressure-
Combustion Portion of Program
-------
CALCULATE THE ENTROPIES 0? THE
PRODUCTS IK -THE BURNING SHELL I
Subroutine ENTR
COMPRESS THE WHOLEH-IIXTURE TO ITS
ORIGINAL VOLUME WITHOUT CHANGES
IN TEMPERATURE AND CONCENTRATION
IN THE SHELLS TO GIVE NEW PRESSURE
Subroutine SOLVE
[CALCULATE THE CONCENTRATION OF |
1 THE PRODUCTS IN THE BURNED SHELLS
AT NEW PRESSURE(AFTER COMPRESSION)
Subroutine CONG
DOES THE
CONCENTRATION
IN ANY OF THE BURNED.
SHELLS CHANGE
'PRECIABLJ
• N
ESTIMATE NFri j
COKCENTRATIOI:!
r
D0\
,
CALCULATE THE ENTROPIES
OF SHELL J AT -NEW PRESSURE
Subroutine ENTR
(ADJUST TEMPERATURE OF
[SHELL J BY ENTROPY
iDEVIATION •
DOES THE
TEMPERATURE OP
ANY SHELL CHANGE
JIABLY2,
Figure A-2 Plow Chart for Equilibrium, Isentropic
Compression Portion of Program
-------
FROM PRESSURE VERSUS TIME
RELATIONSHIP DETERMINE
THE TIME FOR GIVEN PRESSURE
CALCULATE THE."LOWER."LIMIT"
OF .NO. CONCENTRATION IN ;
SHELL i .:.: : •• ••: .. .
Subroutine NO
CALCULATE THE: "UPPER LIMIT"
OF NO CONCENTRATION IN
SHELL I ',. ; .
••"Subroutine NO
: END
Figure A-3 Flow Chart for NO Kinetics Portion of Program
43
-------
Other specific data included within the program are:
(1) the combustion-chamber volume("VT" of main program) in
liters,(2) equilibrium constants ("RK" 's of subroutine "CONG")
determined by curve-fitting of tabulated equilibrium-constant
data from Ref. 28 (3) coefficients for the temperature depend-
ence of enthalpies (constants of subroutine "ENTH") and of
entropies (constants of subroutine "ENTR") drawn from Ref.29,
(4) NO reaction-rate constants ("RK" 's of subroutine "NO"
from Ref. 30 in conjunction with the corresponding equili-
brium constants ("EK" 's of subroutine "NO") from curve-fits
of the data of Ref. 28.
Program Output
Figures A-4, A-5, and A-6 show samples of all possible
program outputs (Code:"MM" = 2) for a fuel-lean octane/air
mixture burning from an initial state at 6.8 atm, 450 K.
Fig. A-4 begins with output showing the format of the
equilibrium output data including: (1) the number of the
shell being described, (2) its temperature(T) and (3) volume
(V), (4) the total no. of moles in that shell (PN) and (5)
the numbers of moles of various species (X's). From these
results, concentrations (moles/mi), or mole fractions are
readily calculated depending on the user's interest.
The numerical equilibrium results are then output for
all burned shells (1 in Fig. A-4, 2 in Fig. A-5) and for
one unburned shell. Since the total chamber volume is sub-
divided by the program into shells of equal size/ the one
unburned shell is typical of all unburned shells.
Finally, kinetic results are displayed, including the
burning time (after ignition), the pressure, and both the
kinetically calculated and the equilibrium NO content of
each burned shell. The average NO contents (kinetic and
equilibrium) of all burned shells is also output. Each
value for NO is in moles. The "overestimated NO" value
given is that calculated for the assumption that NO formation
in any shell begins when that shell begins to burn.
Figure A-5 shows similar results after completion of
burning in two shells out of the total of ten. At this
point and for later calculations an "underestimated NO"
value is output for all previously burned shells. This
value represents the NO formed if it is assumed that each
shell begins forming NO when that shell ends burning. Figure
A-6 shows an example of the kinetic outputfbnly) after com-
pletion of burning by all shells (ten, in this test case).
Figure A-6 is a plot of the kinetic results from the
entire output from which Figures A-4 through A-6 are samples.
44
-------
C 8oOO H 18.00 COMBUSTION
EQUIVALENCE HATIO; 0.90
PRESSURE*ATM): 6-804
TEMPERATURE(K)S 450.00
ARRAY OF OUTPUT DATA
P(ATM)
SHELL T(K)
PN(J>
XH20
XC02CJ>
VCML)
X02CJ) XN2(J>
XHHJ) XOH(J)
XCOCJ) XNOCJ)
XH2CJ)
XOl(J)
EQUILIBRIUM SCHEME
EQUILIBRIUM STATES AFTER COMBUSTION OF SHELL 1
10.95
! 2507-50
0.4145E-02
Oo5048E-03
0*4295E-03
2 508-49
0.3898E-02
OoOOOOE+00
OoOOOOE+00
0.7782362E-01
0.8054E-04 0.3033E-02
0.1310E-05 0.21I4E-?04
0.3521E-04 0.2972E-04
0-1484215E-01
0.8067E-03 0-3033E-02
O.OOOOE+00 0-OOOOE+OO
OoOOOOE+00 O.OOOOE+00
0.6776E-05
Oo2606E-05
OoOOOOE+00
0-OOOOE+OO
KINETIC SCHEME
OVERESTIMATED NO
SHELL TIME INTERVAL
1 Oo260E+02
TIME(MS)=26-0
PCATM)= 10«95
NO CONCENTRATION
0.29E-04
EQUILIBRIUM NO
0-30E-04
AVERAGE* 0«29E-04 AVERAGE= 0«30E-04
Figure A-4
Partial Sample Output from Program -
Equilibrium and Kinetic Results After
Burning of 1st Shell
-------
EQUILIBRIUM SCHEME
EQUILIBRIUM
14o91
1 2636.53
Oo4163E-02 0
0»4971E-03 0
Oo/jlllE-03 0
2 2482..81
Oo4138E-02 0
Oo5080E-03 0
Oo4373E-03 0
3 ,550»87
Oo3898E-02 0
OoOOOOE+00 0
OoOOOOE+00 0
STATES AFTER COMBUSTION OF SHELL 2
0.6038301E-
.8496E-04
o2355E-05
o5361E-04
0»5651675E-
o7778E-04
•8870E-06
o2742E-04
0-1181291E-
•8067E-03
oOOOOE+00
oOOOOE+00
01
Oo3033E-02
0.2882E-04
0.3770E-04
01
0.3033E-02
0»1801E-04
0«2798E-04
01
0»3033E-02
O.OOOOE+00
0»OOOOE+00
Ool011E-04
0-4198E-05
Oo5269E-05
0«1941E-05
0*OOOOE+00
OeOOOOE+00
KINETIC SCHEME TIME38E-04
2 Oo573E-»-dl Ool8E-04 0»28E-Q4
AVERAGED 0«27E-04 AVERAGE= Oo33E°04
Figure A-5
Partial Sample Output from Program -
Equilibrium and Kinetic Results After
Burning of 2nd Shell
46
-------
KINETIC SCHEME TIME(MS)=54.0 PCATK>= 44.48
UNDERESTIMATED MO(COMPARE WITH LAST OVERESTIMATED NO CALCULATION)
SHELL TIME INTERVAL NO CONCENTRATION EQUILIBRIUM NO
1 0»194E+00 0.74E-04 0«75E-04
8 0.479E+00 0«58£-04 Q.59E-04
3 Oo906E+00 0.50E-04 0»50E-04
4 0.150E+01 0.44E-04 0.44E-04
5 Oa287E+01 0.39E-04 0-40E-04
6 0.414E+01 0.36E-04 0.36E-04
7 0..558E+01 0.33E-04 0.34E-04
8 Oo558E+01 Oo30E-04 0.31E-04
9 Oo558E+01 0.26E-04 Oo29E-04
AVERAGE= 0.39E-04 AVERAGE= 0»40E-04
OVERESTIMATED NO
SHELL TIME INTERVAL NO CONCENTRATION EQUILIBRIUM NO
1 Ool51E+00 0.78E-04 0.79E-04
2 0.363E+00 0«62E-04 0-62E-04
3 0«678E+00 0.53E-04 Oc53E-04
4 OolllE+01 0.46E-04 0.47E-04
5 0-168E+01 Oo4SE-04 0«42E-04
6 Oo240E-«-01 0«38E-04 0.39E-04
7 0»/I12E + 01 0.36E-04 0.36E-04
8 Oo558E+01 0.33E-04 O
9 0«558E+01 0-31E-04 O
10 Oo558E+01 0.27E-04 0-30E-04
AVERAGE= 0.45E-04 AVERAGE= 0.45E-04
Figure A-6 Partial Sample Output from Program -
Kinetic Results Only, After Burning
of 10th (Last) Shell
-------
20
n
I
o
Du
O<
I 10
2
W
u
2
O
U
CoH-,0 / AIR (STOICHIOMETRIC)
o lo
INITIAL CONDITIONS:
6.8 ATM ,
LOWER LIMIT -—
SHELL JUST BURNED
4
2 341!
f 1 I ll
5 6 7 8 9
I I I I
1C
10
20 30
TIME (msec)
40
50
Figure A-7 Calculated Upper and Lower Limits of
NO Concentration
48
-------
Program Checks
Program output was checked, so far as practical, by
comparison of results with those from a more comprehensive
thermochemical program. The program of Reference 29 (here-
after termed: "reference program") was available for this
purpose in the Combustion Laboratory computer-program
library. Comparisons were made for two cases.
First, the results of the present program for constant-
pressure combustion of a single shell were compared with
those from the reference program. This uncovered several
shortcomings in the original form of the present program
[23], and it was modified accordingly. Results from the
two programs agreed very closely.
Second, the calculated total entropy and composition
from the present program for constant-volume combustion
of a single shell were compared with those calculated by
the reference program. Input to the reference program
were temperature and pressure after combustion output by
the present program. The reference program calculations
for composition and total entropy were in excellent agree-
ment with those of the present program.
-------
DIMENSION T(50)J.VC50),PN(50).,Fr\K50),X02(50),XN2(50),ENTC50),
4 PML ( 5 0 ) > ENT.V C 50 ; > XCO 2 C 5 0 > » XCO ( 50 ) > XH2 ( 50 ) > XH 20 C 50 ) * XH 1 ( 5 0 > >
5 XOH (50), XO 1(50), TIN ( 50 ) > XfcO ( 50 ) * YiNJO ( 50 ) , PC ( 20 ) , TE ( 20 ) *
X TF( 10>,VF( 10)^X02F( 1C).*XNOF( 10)*X01F( 1 0) ,XN2FC 1 0) *
X XOHF(10)*XH1F(10)
DATA PHI j.p*TI/0. 9,6. 804/450. /
DATA CARB^HYD/8.^18./
DATA NS/NPT/10, 1 2/
DATA B1*B2>B3/-58370« *7.*.03/
DATA Cl^C2*C3/52. 89*70, «06/
DATA MM/2/
DATA 0
D'ATA
1 44«*48.*52.*54./
NPT=NPT-1
80 XNUM=CARE+HYD/4.
PIN=P
VT=« 21 1/403
WRITE(6,30)CARB,HYD,PHI
30 FORMAT(3HIC F4.2,3H H F5.2, 12H COMBUSTION,/
X 20HOEQU I VALENCE RATIO: F4.2)
31 FORMATC16H PfiESSUREC ATM) '. F7»3* 10X* 1 7H TEMPERATURE ( K) : ,F7-2//
TN=P*VT/( .082+TI )
FNT=TN*PHI / C PHI +XNUM*4 . 76 )
RNS=NS
XC02(0)=(CARB-. 1 )*FNT/RNS
XH20(0)=(HYD/2.-. 1 )*FNT/RNS
XT02=XNUM/PHI *FNT
XTN2=3-76*XT02
-P02=XT02/TN*P
PN2=XTN2/TN*P
PF«=FNT/TN*P
CALL ENTRCl.pC2*C3>
ENTT=FNT*SF+XTN2*SN2+XT02*S02
DO i I^IJ-NS
T(I)=TI
VCI)=VT/RMS
PNU)=TN/RNS
FN(I)=FNT/RMS
XN2CI )=XTNP/RNS
X02
-------
WRITEC6*14>
14 FORMATC ARRAY OF OUTPUT DATA'/' P(ATM)'/
X ° SHELL%2Xj>' T(K)°P10X*' VCMLV/SX*' PN%
X 6X* ° X02CJ>'»5Xj' XN2CJ)'j5X«' XH2(J)V5X*
X ' XH20%4X,» XH1CJ) ',5X> ' XOH< J> ' * 5X* ' . XOKJ)1
X /5X, ' XC02(AJ)',4X*° XCO(J)'*5X>' XNOCJ)'//)
98 DO 2 I=1*NS
CALL ENTHCTU>,EF*E02,EH2,EN2,EH20,EHUEOH,E01*EC02,
X ECO*B1*B2*B3) .
ENF=FN(I>*EF
EN02=X021>
21 CALL CONC(T(I)j>P*PHIj.PN(I),FN(I)*XH20(I)*XH2(I)*XHl(I)*XOH(I)*
6 XC02CI >»XCOCI > ;XN2CI )*XG2CI >*X01 CI ) ,XNOC I ) ^CARD
CALL ENTK(T(l)*EF*EOa*EH2*EN2*EH20*EHl*EOH*E01*EC02>
X ECO/.B1*B2*B3)
EN02=E02*X02(I>
ENN2=EN2*XN2CI)
ENH2=£H2*XH2U>
ENH20=EH20*XH20( I)
ENH1=EH1*XHKI)
ENC02=EC02*XC02( I >
ENCO=ECO*XCO(I)
ENOH=EOH*XOHU>
EN01=E01*X01(I)
ENTHN=EN02+ENN2+ENH2+ENH20+ENHl+ENC02+ENCO+ENOH-»-ENdl
TBN=T(I)-KENTHT-ENTHN)/»16
IF(ABS(TBN-T(I».LT..0005*TCI))GO TO 20
GO TO 21
20 CONTINUE
P02=X02(I)/PNU)*P
J%M2=XN2(I)/FNCI)*P
PH2=XH2CI)/PN(I)*P
PH20=XH20CI )/PN(I)*P
PH1«=XH1(I)/PN(I>*P
POH=XOH< I )/PNCI ),*P
P01=X01 (I)/PN*P
PC02=XC02.*P
FOKMAT(////19H EQUILIBRIUM SCHEME*/)
•IFCMM.EQ.3«'ANP..I.Ee.l)60 TO 40
GO TO 51
40 VRITECej-SCnPj.IjTCDjPNCI ),X02(I > ,XN2( I 5 , XH2( I )
1 XHKI >,XOHJ.XC02CI),XCOC:>J>XNOCI)
50 . FORMAT(29H CONSTANT PRESSURE COMBUSTION,/*
1 lX>F6.2/< 1XJ.I3,2X*F7«2/4X»4E12.4/4X*4E12.4/4X*3E12.4))
51 CALL ENTfiCTCI ) *PF*P02,PH2J.?N2,PH1 pPO 1 *PC02>PCO*PH20*POH* SF*
51
-------
5 S02»SH2,SN2*SH1,S01 *SC02*SCO*SH20»SOH*C1*C2*C3>
S02=X02(I)*S02
SN2=XN2( I )*SN2
SH2-XH2U )*SH2
SH1=XH1(I )*SH1
S01=X01(I >*S01
SC02=XC02U)*SC02
SCO=XCOCI >+SCG
SH20=XH20CI )*SH20
SOH=XOHCI >*-?OH
ENTCI )=S02+SN2-t-SH2+SHl+S01+SC02+SCO + SH20 + SOH
CALL SOLVE(VTj.NS>P*T,FN*V>
K=10
L=10
DO 3 J~lj>l
PNL(J)=FW(J)
CALL CONC(T(J)*P^PHI*PN(J)*FN*XNOCJ)*CARB*HYD*XNUM)
E}0 4 J= 1 * I
^F+PNLCJ))/2«
K=100
CONTINUE
IF(KoEQ»100)GO TO 5
DO 6 J=1,NS
P02=X02CJD/PN(J)*P
PN2=XN2(J)/PN( J)*P
PH2=XH2(J)/PN(J)*P
P01=XOKJ)/PNCJ)*P
PH1=XH1(J)/PNCJ)*P
PC02=XC02( J)/PNCJ)*P
PCO=XCOC J)/PNCJ)*P
PH20=XH20CJ)/PN(J)*P
POH=XOH C J ) /PN C J) *P
PF=FN(J)/PN( J)*P
CALL ENTRCTC J)*PF*P02*PH2,PN2^PHlJ.P01*PC02*PCO*PH20*POHx.SFJ,
IF(JoGT<.I)SF=FNCJ)*SF
S02=X02CJ)*S02
SN2=XN2( J)*SN2
SH2=XH2(J)*SH2
SH1=SH1*XH1(J)
S01=S01*XOKJ>
SC02=SC02*XC02(J/
SCO=SCO*XCOCJ)
SH20=SH20*XH20(J)
SOH=SOH*XOHCJ)
IFCJoGToDGO TO 7
ENTN(J)=S02-»-S.M2+SH2+SHl+S01-»-SC02-«-SCO+SH20-»-SOH
GO TO 6
7 ENTN=SN2+S02+SF
6 TTN(J)=T(J)*EXPC(ENT(J)-ENTN(J))/«19)
52
-------
DO 8 J=1.»N3
IrCABSCTTNC J)-T( J.» .LT« .0005*7 ( J))GO TO 3
T( J)-(TTH( J>+T< J>>/2.
L=100
8 CONTINUE
IFCL-EQ* 1COGO TO 5
1FCMM.E3. 1 )GO TO 1049
IFCI.EQoNS>GO TO 279
11=1+1
GO TO 379
279 II=NS
379 WSI7E(6*60H
60 FORMATC EQUILIBRIUM STATES AFTER COMBUSTION OF SHELL' 1 3)
WRITEC6,70>Pj.C J/T( J),V( J> *?N( J>>X02( J),X*:2C J).»XH2C J>,
X XH20C J),XH1 ( J^XO:i( J)*XDl(o)^XC02( J)*XCO( J
70 FORMAT ( lX*F6.2/( 1 X* I 3* 2X> F7. 2*4X>E/4X*4E 12
X 4El2./i/^X>3E12.^i) )
!F*SLOPE
DTIME=TIME-TIML '
WRITEC6/69)TIME^P .
69 FORMAT(/15H KINETIC SCHEME^SX*1 TIKE*
X XN2< J)*XOHC J)J»XH1 f J),TF< J)j»VFC J)j.X02F(J) »XNOF( J)>
X X01F(J)*XN2F( J)*XOHFCJ)jXHlF( J)
SUMF=SUMF+YNO< j)
SUMMOF=SUMNOF+XNOFCJ)
7^ CONTINUE
RKK=KK
IFU.EQ.NS)RKK=RNS
AVGF=SUMF/RKK
53
-------
AVGNOF=SUMNOF/RKK
WRITEC6,61>AVGF,AVGNOF
61 FORMATC21X,' AVERAGE= ',E8.2*11H AVERAGE«= *E8»2>
C OVERESTIMATED NO
75 KODE=2
WRITEC6*76)
76 FORMATC OVERESTIMATED NOV6H SHELL*4X*
X 13HTIME INTERVAL*4Xj> 16HNO CONCENTRATION*4X*
X 14HEQUILIBRIUM WO)
KK=I
SUKB=O.O
SUMNOB=O.O
DO 77 J=1*KK
CALL NO(PHI,KODE*T(J)*VCJ)*YNO(J)*X02(J)*XNO*TF(J)*VF(J)*X02F(J»XNOF(J)*
X X01F(J)*XN2F(J)*XOHF(J)*XH1FAVGB,AVGNOB
62 FORMATC21X, * AVERAGE" '*E8.2*HH AVERAGE- ,E8»2>
DO 78 J=1*I
TF(J)=T(J)
VF(J)=V
XN2F
-------
SUBROUTINE CONC(T*P*PHI J>FN,FN*H20*H2J,H1 *OK^C02*CO*RW2*02* 011.
X RNO*CARBpHYD,XNUM>
CDIF=OoO
ODIF=0»0
HDIF=OoO
1 PNL=PN
CON=P/PN
RK1 = 10 «**<<• 1.321455E05/T-«3061947E01>*SQRT(CON>
RK2 = 10.**(!r- 1087288E04/T+. 1219451 1 1E01 )
RK3 = 10.**('«30l3006E05/T-o75131088E01)*CON
RK5 = 10o**(;-.2395742E05/T+.6380064E01 )/CON
RK6=10o**(-o2679.026E05/T + «7031926EOl )/CON
RK8=10o**(-o4701074E04/T*.6538625EOO)
0=FN*2«*XNUM/PHI
C°FN*CARB
H=FN*HYD
\H IFGO TO 2
303 C02=C*02**o5*RKl/(P.K2+02**»5*RKl)
2 H2a(C-C02)*H20/C;RK2*C02)
H2=OoOO
Hl=«RK5*(C-C02)*H20/(RK2*C02))**(.5)
5 Hl^OoOO
6 OHs10«-ll»ll
10 OHL=OoOO
11 CO=C-C02
12
15 IFCPHI.LT. l.OJGO TO 17
02°CRK2*C02/CRK1*
-------
42 RNO=0.00
C 0 ATOM BALANCE
19 ON=2o*C02+CO+H20+OHL+2.*02+01+RNO
IF0:GO TO 300
C02L=C02
C02=C02-(ON-0)/2«>
CDIFL=CDIF
CDIF=C02-C02L
IFCABSCCDIF>.LE-.5<.E-8>GO TO 43
IF(CDIFL.EQ.O.O)GO TO 14
IF(CDIF/CDIFL>2010>43, 14
2010 C02=CC02+C02L>/2»
GO TO 14
300 02L=02
ODIFL=ODIF
ODIF=02-02L
IF«ABS*PN<50>
SUM=0«0
DO 100 1=1,N
100 SUM=SUM+PNCI)*o082*T(I>
P=SUM/VT
DO 101 1=1,N
101 VSCI)=PN(I)*.082#T(I)/P
RETURN
END
-------
SUBROUTINE t?JTH( T*Er »E02*EK2»EN2*EH20*EH 1 ^EOH,EO 1 *ECQ2,
6 E/5o*T**5+F)
IF(T»GTo 1000»)GO TO 50
E02=H(TJ>3.7189946,-2.5l67238E-03J.8.5e37353E-06>
8 -8o2S98716E-09*2.7082180E-12,-1.0576706S03)
EN2*H(T*3«6927389p-1.34l 1998E-03p2i6693385E-06j.
7 -9
qo TO si i
50 E02=HCT*3<.5980995j>7.8056719E-04J.-2.fi337506E-07,
8 4»2396622E-1 1 »-3. 34 1 9056E- 15*- 1 • 1929753E03)
EN2=HCT> 2. 8609546* 1 .565 1 765E-03«-S. 1 7799O1E-Q7;
9 lol
6 .-3.4633889E-il«3o6719053Z-15«-8.68e20'51£02>
EH80«'H-'8.801 7921E-09*
7 5o9643621E-12>-5.5743608E-16*2.9230007E04)
EC02*H-1.7094645E-14>-4.5940097EOA>
ECO=H-7o7208967E-l5j>-lo42328b7E04>
51 RETURN
END
•-, /
57
-------
SUBROUTINE ENTJJC Tj>PF.»P02.»PH2j»PN2.,PHl >POi >PC02*PCOj»PH20*
2 *X**3+E/4«*X**4+F-ALOGCY))
IF-2.5167288E-03J.8.5837353E-06J>
4 -8«2998716E-09,2.70S2180E-12*3o9080704)
SN2=SCT,PM2J. 3. 6927389*-!. 34 11998E-03, 2. 6693385E-06*
7 -9«.9559574E-10*-9.3276676E-14,2.2827699)
SF=CH-C2*ALOG
GO TO 201
200 SC02=S(T^PC02^4e4008955^3.2163852E-03^-l .3134666E-06J.
5 2<>4548216E-10>-lo7094645E-14*-6. 6437717E-01 )
5 1.1272797E-10J-7.7208967E-15>6. 5172790)
SH20=SPK20, 2- 6707532^3. 031 71 15E-03, -8»5351 570E-07,
5 lol790853E-10*-6.1973568E-15*6.8838391)
5 -3« 4633889E-11,3.6719053E-15*-1« 8472212)
7 -8.«3272843E-13.,7.1462541E-17,-4o7447219E-01)
SOH=S(T*POH*2.9029511J.9«7345968E-04*-2.0321080E-07j.
7 1 .59 1 161 2E-11, -5. 1673895E- 17/5. 4854499)
S02=SCT*P02* 3 -5980995*7o 8056 71 9E-04, -2. 233 7506E-07*
7 4»2396622E-1 lj.-3.3419056E-15*3. 7465771 )
SOi=S -5. 5 74360SE- 16*4. 9467942)
SN2=SCT*PN2* 2. 8609546*1. 585 1765E-03* -6«1 77990 1E-07,
5 1»1119420E-10*-7»5210614E-15>6. 3553564)
201 RETURN
END
58
-------
SUBROUTINE NO (PHI *KODE*T* V* YNO*X02,XNO,X01 *XN2*XOH*
X XHlJ.TF;.VF*X02F,XNOF,X01F,XN2F,XOHF*XHlF*TIME.i>
X TIMLjJ*KK*I*DT.rME>
DIMENSION ZMO('10«2)
IFCKODE«EQol)GO TO 10
TY=T
VY=V
-• X02Y=X02
X01Y=X01
XN2Y=XN2
XOrtY=XOH
»liY=XHl
GO TO 1 1
10 TY=TF
VY=VF
XNOY=XNOF
X02Y=X02F
X01Y=X01F
XN2Y=XN2F
XOHY=XOHF
XH1Y=XH1F
U VM=1000«*VY
ZNOE=XNOY/VM
Z01=X01Y/VM
ZN2=XN2Y/VM
ZOH=XOHY/VM
Z02=X02Y/VM
ZH1=XH1Y/VM
RK2=6o4E09*TY*EXP(-3140«/TY)
EKl=10o**C-o 1 6480 79^E05/TY+. 6723525)
EK2=10»#*<« 70 78646EO^/TY+. 6353725)
EK3=10.**(.103^5546E05/TY-.3851659)
B=RK1/EK1
IF
-------
1200
1250
1260
1300
2111
GO TO 1250
ZNOL=O«O
ZNOCJ*KODE)=CZNOE+ZNOL>/2.
E=SQRT(C/D)
H=ALOGCE+ZWOL)
R=ALOGCE-ZNOL)
F=ALOG(E + ZNO( J*KODE»
G=ALOG CE-ZNO C J, KODE >)
TIMEW=TIML+ 100C'«.*CA*(F+R-G-H)/(2.*D*E)-B*(F+G-R-H)/C2.*D)
IFCABS(TIME-TIMEW).LE. .001*DTIME)GO TO 2111
•IF(ZNO(J*KODE).GE..99*ZNOE«AND.TIMEW.LT.TIME)GO TO 2111
DER=1000..*CA+E*ZNO( J> KODE) )/ CD* GO TO 1260
ZNO C J, KODE ) =ZNO ( J^ KODE ) - C T IME-TI MEW ) /DER
ZNO ( J* KODE )=( ZNO ( J* KODE ) +ZNOE ) /2 .
GO TO 1260
YNO=ZNO(J*KODE)*VM
XNOE=ZNOE*VM
DTIK=TIMEW-TIML
RETURN
END
C, 9.
-------
TECHNICAL REPORT DATA
(Please read Instructions oh the reverse before completing)
1. REPORT NO.
APTD-1498
3. RECIPIENT'S \CCESSIOWNO.
4. TITLE AND SUBTITLE
5. REPORT DATE
Nitric-Oxide Measurement in a Simulated Spark-Ignition
Engine
January 1975
6. PERFORMING ORGANIZATION CODE
7,_A(JTHOR
-------
INSTRUCTIONS
1. REPORT NUMBER
Insert the EPA report number as it appears on the cover of the publication.
2. LEAVE BLANK
3. RECIPIENTS ACCESSION NUMBER
Reserved for use by each report recipient.
4. TITLE AND SUBTITLE
Title should indicate clearly and briefly the subject coverage of the report, and be displayed prominently. Set subtitle, if used, in smaller
type or otherwise subordinate it to main title. When a report is prepared in more than one volume, repeat the primary title, add volume
number and include subtitle for the specific title.
5. REPORT DATE
Each report shall carry a date indicating at least month and year. Indicate the basis on which it was selected (e.g., date of issue, date of
approval, date of preparation, etc.).
6. PERFORMING ORGANIZATION CODE
Leave blank.
7. AUTHOR(S)
Give name(s) in conventional order (John R. Doe, J. Robert Doe, etc.). List author's affiliation if it differs from the performing organi-
zation.
8. PERFORMING ORGANIZATION REPORT NUMBER
Insert if performing organization wishes to assign this number.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Give name, street, city, state, and ZIP code. List no more than two levels of an organizational hircarchy.
10. PROGRAM ELEMENT NUMBER
Use the program element number under which the report was prepared. Subordinate numbers may be included in parentheses.
11. CONTRACT/GRANT NUMBER
Insert contract or grant number under which report was prepared.
12. SPONSORING AGENCY NAME AND ADDRESS
Include ZIP code.
13. TYPE OF REPORT AND PERIOD COVERED
Indicate interim final, etc., and if applicable, dates covered.
14. SPONSORING AGENCY CODE
Leave blank.
15. SUPPLEMENTARY NOTES
Enter information not included elsewhere but useful, such as: Prepared in cooperation with, Translation of, Presented at conference of,
To be published in, Supersedes, Supplements, etc.
16. ABSTRACT
Include a brief (200 words or less} factual summary of the most significant information contained in the report. If the report contains a
significant bibliography or literature survey, mention it here.
17. KEY WORDS AND DOCUMENT ANALYSIS
(a) DESCRIPTORS - Select from the Thesaurus of Engineering and Scientific Terms the proper authorized terms that identify the major
concept of the research and are sufficiently specific and precise to be used as index entries for cataloging.
(b) IDENTIFIERS AND OPEN-ENDED TERMS - Use identifiers for project names, code names, equipment designators, etc. Use open-
ended terms written in descriptor form for those subjects for which no descriptor exists.
(c) COSATI HELD GROUP - Field and group assignments are to be taken from the 1965 COSATI Subject Category List. Since the ma-
jority of documents are multidisciplinary in nature, the Primary Field/Group assignment(s) will be specific discipline, area of human
endeavor, or type of physical object. The application(s) will be cross-referenced with secondary Field/Group assignments that will follow
the primary posting(s).
18. DISTRIBUTION STATEMENT
Denote releasability to the public or limitation for reasons other than security for example "Release Unlimited." Cite any availability to
the public, with address and price. ,
19. &20. SECURITY CLASSIFICATION
DO NOT submit classified reports to the National Technical Information service.
21. NUMBER OF PAGES
Insert the total number of pages, including this one and unnumbered pages, but exclude distribution list, if any.
22. PRICE
Insert the price set by the National Technical Information Service or the Government Printing Office, if known.
EPA Form 2220-1 (9-73) (Reverse)
------- |