EPA 600/3 75-010d
September 1975
Ecological Research Series
ANNUAL CATALYST RESEARCH PROGRAM REPORT
APPENDICES
Volume III
lealth Effects Research Laboratory
Office of Research and Development
U.S. Environmental Protection Agency
Research Triangle Park, N.C. 27711
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development,
U.S. Environmental Protection Agency, have been grouped into
five series. These five broad categories were established to
facilitate further development and application of environmental
technology. Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in
related fields. The five series are:
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental Studies
This report has been assigned to the ECOLOGICAL RESEARCH series.
This series describes research on the effects of pollution on
humans, plant and animal species, and materials. Problems are
assessed for their long- and short-term influences. Investigations
include formation, transport, and pathway studies to determine the
fate of pollutants and their effects. This work provides the
technical basis for setting standards to minimize undesirable
changes in living organisms in the aquatic, terrestrial and
atmospheric environments.
This document is available to the public through the National
Technical Information Service, Springfield, Virginia 22161.
-------
EPA-600/3-75-010d
September 1975
ANNUAL CATALYST RESEARCH PROGRAM REPORT APPENDICES
Volume III
by
Criteria and Special Studies Office
Health Effects Research Laboratory
Research Triangle Park, North Carolina 27711
U.S. ENVIRONMENTAL PROTECTION AGENCY
OFFICE OF RESEARCH AMD DEVELOPMENT
HEALTH EFFECTS RESEARCH LABORATORY
RESEARCH TRIANGLE PARK, NORTH CAROLINA 27711
-------
CONTENTS
Page
CATALYST RESEARCH PROGRAM ANNUAL REPORT
EXECUTIVE SUMMARY 1
INTRODUCTION 5
PROGRAM SUMMARY 7
TECHNICAL CONCLUSIONS 17
DISCUSSION 22
REFERENCES 45
APPENDICES TO CATALYST RESEARCH PROGRAM ANNUAL REPORT
VOLUME 1
A. OFFICE OF AIR AND WASTE MANAGEMENT
A1. AUTOMOTIVE SULFATE EMISSIONS 1
A2. GASOLINE DE-SULFURIZATION - SUMMARY 53
A2.1 Control of Automotive Sulfate Emissions
through Fuel Modifications 55
A2.2 Production of Low-sulfur Gasoline 90
VOLUME 2
B. OFFICE OF RESEARCH AND DEVELOPMENT
B1. FUEL SURVEILLANCE
B1.1 Fuel Surveillance and Analysis 1
B1.2 The EPA National Fuels Surveillance
Network. I. Trace Constituents in Gasoline
and Commercial Gasoline Fuel Additives . . 19
B2. EMISSIONS CHARACTERIZATION
B2.1 Emissions Characterization Summary .... 44
B2.2 Sulfate Emissions from Catalyst- and Non-
catalyst-equipped Automobiles 45
B2.3 Status Report: Characterize Paniculate
Emissions - Prototype Catalyst Cars .... 68
B2.4 Status Report: Characterize Paniculate
Emissions from Production Catalyst Cars . . 132
B2.5 Status Report: Survey Gaseous and Particu-
late Emissions - California 1975 Model Year
Vehicles 133
B2.6 Status Report: Characterization and Meas-
urement of Regulated, Sulfate, and Particu-
late Emissions from In-use Catalyst Vehicles -
1975 National Standard 134
B2.7 Gaseous Emissions Associated with Gasoline
Additives - Reciprocating Engines. Progress
Reports and Draft Final Report - "Effect of
Gasoline Additives on Gaseous Emissions" . 135
-------
Page
B2.8 Characterization of Gaseous Emissions from
Rotary Engines using Additive Fuel -
Progress Reports 220
B2.9 Status Report: Exploratory Investigation of
the Toxic and Carcinogenic Partial Combus-
tion Products from Oxygen- and Sulfur-
containing Additives 232
B2.10 Status Report: Exploratory Investigation of
the Toxic and Carcinogenic Partial Combus-
tion Products from Various Nitrogen-
containing Additives 233
B2.11 Status Report: Characterize Diesel Gaseous
and Particulate Emissions with Paper "Light-
duty Diesel Exhaust Emissions" 234
B2.12 Status Report: Characterize Rotary Emissions
as a Function of Lubricant Composition and
Fuel/Lubricant Interaction 242
B2.13 Status Report: Characterize Particulate
Emissions - Alternate Power Systems
(Rotary) 243
VOLUME 3
B.3 Emissions Measurement Methodology
B3.1 Emissions Measurement Methodology Summary 1
B3.2 Status Report: Develop Methods for Total
Sulfur, Sulfate, and other Sulfur Compounds
in Particulate Emissions from Mobile Sources 2
B3.3 Status Report: Adapt Methods for SO2 and SO3
to Mobile Source Emissions Measurements 3
B3.4 Evaluation of the Adaption to Mobile Source
SO2 and Sulfate Emission Measurements of
Stationary Source Manual Methods 4
B3.5 Sulfate Method Comparison Study. CRC APRAC
Project CAPI-8-74 17
B3.6 Determination of Soluble Sulfates in CVS
Diluted Exhausts: An Automated Method 19
B3.7 Engine Room Dilution Tube Flow Characteristics. ... 41
B3.8 An EPA Automobile Emissions Laboratory 52
B3.9 Status Report: Protocol to Characterize Gaseous
Emissions as a Function of Fuel and Additive
Composition - Prototype Vehicles 89
B3.10 Status Report: Protocol to Characterize Particu-
late Emissions as a Function of Fuel and Additive
Composition 90
B3.ll Interim Report and Subsequent Progress Reports:
Development of a Methodology for Determination
of the Effects of Diesel Fuel and Fuel Additives
on Particulate Emissions 192
-------
Ptigc
B3.12 Monthly Progress Report #7. Protocol to
Characterize Gaseous Emissions as a Function
of Fuel and Additive Composition 200
B3 13 Status Report Validate Engine Dynomometer Test
Protocol for Control System Performance 218
B3.14 Fuel Additive Protocol Development 221
B3.15 Proposed EPA Protocol- Control System
Performance 231
VOLUME 4
B3.16 The Effect of Fuels and Fuel Additives on Mobile
Source Exhaust Particulate Emissions 1
VOLUME 5
B3.17 Development of Methodology to Determine the
Effect of Fuels and Fuel Additives on the Perform-
ance of Emission Control Devices 1
B3.18 Status of Mobile Source and Quality Assurance
Programs 260
VOLUME 6
B4. Toxicology
B4.1 Toxicology: Overview and Summary 1
B4.2 Sulfunc Acid Effect on Deposition of Radioactive
Aerosol in the Respiratory Tract of Guinea Pigs,
October 1974 38
B4.3 Sulfuric Acid Aerosol Effects on Clearance of
Streptococci from the Respiratory Tract of Mice.
July 1974 63
B4.4 Ammonium and Sulfate Ion Release of Histamine
from Lung Fragments ' 89
B4.5 Toxicity of Palladium, Platinum and their
Compounds 105
B4.6 Method Development and Subsequent Survey
Analysis of Experimental Rat Tissue for PT, Mn,
and Pb Content, March 1974 128
84.7 Assessment of Fuel Additives Emissions Toxicity
via Selected Assays of Nucleic Acid and Protein
Synthesis 157
B4.8 Determination of No-effect Levels of Pt-group
Base Metal Compounds Using Mouse Infectivity
Model, August 1974 and November 1974 (2
quarterly reports) 220
B4.9 Status Report- "Exposure of Tissue Culture
Systems to Air Pollutants under Conditions
Simulating Physiologic Stales of Lung and
Conjunctiva" 239
-------
Page
B4.10 A Comparative Study of the Effect of Inhalation of
Platinum, Lead, and Other Base Metal Compounds
Utilizing the Pulmonary M^rophage as an Indicator
of Toxicity 256
B4.11 Status Report: "Compare Pulmonary Carcinogenesis
of Platinum Croup Metal Compounds and Lead Com-
pounds in Association with Polynuclear Aromatics
Using hi vivo Hamster System 258
B4.12 Status Report: Methylation Chemistry of Platinum,
Palladium, Lead, and Manganese 263
VOLUME 7
B.5 Inhalation Toxicology
B5.1 Studies on Catalytic Components and Exhaust
Emissions 1
B.6 Meteorological Modelling
B6.1 Meteorological Modelling - Summary 149
B6.2 HIWAY: A Highway Air Pollution Model 151
B6.3 Line Source Modelling 209
B.7 Atmospheric Chemistry
B7.1 Status Report: A Development of Methodology to
Determine the Effects of Fuel and Additives on
Atmospheric Visibility 233
Monthly Progress Report: October 1974 255
B7.2 Status Report: Develop Laboratory Method for Collec-
tion and Analysis of Sulfuric Acid and Sulfates • • • - 259
B7.3 Status Report: Develop Portable Device for Collection
of Sulfate and Sulfuric Acid 260
B7.4 Status Report: Personal Exposure Meters for
Suspended Sulfates 261
B7.5 Status Report: Smog Chamber Study of SO,
Photo-oxidation to SO under Roadway
Condition 262
B7.6 Status Report: Study of Scavenging of SO. and
Sulfates by Surfaces near Roadways . . . 263
B7.7 Status Report: Characterization of Roadside
Aerosols: St. Louis Roadway Sulfate Study 264
B7.S Status Report: Characterization of Roadside
Aerosols: Los Angeles Roadway Sulfate Study • • • • 269
-------
Page
VOLUME 8
B.8 Monitoring
B8.1 Los Angeles Catalyst Study. Background Pre-
liminary Report
B8 2 Los Angeles Catalyst Study; Summary of Back-
ground Period (June, July, August 1974)
B8.3 Los Angeles Catalyst Study Operations Manual
(June 1974, amended August 1974)
B8 4 Collection and Analysis of Airborne Suspended
Particulate Matter Rcspirablc to Humans for
Sulfales and Polycyclic Organics (October 8, 1974).
1
13
33
VOLUME 9
B 9 Human Studios
1974
.194
1
B9.1 Update of Health Effects of Sulfates, August 28,
B9 2 Development of Analytic Techniques to Measure
Human Exposure to Fuel Additives, March 1974 .... 7
B9.3 Design of Procedures for Monitoring Platinum
and Palladium, April 1974 166
B9 4 Trace Metals in Occupational and Non-occupation-
ally Exposed Individuals, April 1974 178
B9.5 Evaluation of Analytic Methods for Platinum and
Palladium .' 199
B9.6 Literature Search on the Use of Platinum and
Palladium 209
B9.7 Work Plan for Obtaining Baseline Levels of Pt
and Pd in Human Tissue 254
-------
Appendix B3.1
Emissions Measurement Methodology Summary
The major effort in the ORD Fuel and Fuel Additive, Catalyst,
and Mobile Source Emissions Research Programs has involved detailed
characterization of non-regulated gaseous and particulate pollutants
from mobile sources. The effect of fuel composition, fuel additives,
and control devices have been the predominant determinants of interest.
Appendix B2 covered details of the emissions characterization repeats
of the program. Within this appendix, those programs specifically
directed toward development of standardized measurement methods to be
routinely used for the assessment of particulate, sulfate and S02
emissions from non-catalyst and catalyst equipped vehicles will be
reviewed.
-------
Appendix B3.2
Status Report
ROAR 21BCE
Task 043
Develop Methods for Total Sulfur. Sulfate, and
Other Sulfur Compounds in Parti oil ate Emissions from Mobile Sources
This in-house project is designed to provide the methodology
necessary to analyze the particulate matter collected from mobile
source emissions. The fate of organic sulfur compounds both present
in fuel and added in additive packages has not been fully defined.
While a sulfate method has been developed, other methods are needed
for total sulfur and other inorganic and organic sulfur compounds.
X-ray fluorescence methods are being investigated to measure total
sulfur. Instrumental methods to permit the analysis of sulfuric acid
in addition to the method previously developed may be evaluated. Other
inorganic sulfides, sulfites, as well as organic sulfur compounds, will
be investigated on the basis of probable impact.
Status:
Recruitment is under way and the project should be implemented in
the second quarter of FY75.
-------
Appendix B3.3
Status Report
ROAP 21BCE
Task 042
Adapt Methods for SO? and SO^ to
Mobile Source Emissions Measurements
This in-house project will further implement and provide more
definitive efforts to develop methodology applicable to use in mobile
source analyses. Orginally,efforts have provided exploratory methodology
to determine sulfur dioxide concentrations in mobile source emissions.
This effort will be expanded to provide better methods and to correlate
these with other methodology. The need is for real time monitors for
gaseous sulfur emissions which may be used to monitor systems in operation,
Status:
Employment of appropriate manpower is under way and should be
completed shortly allowing activation of this project.
-------
Appendix B3.4
EVALUATION OF THE ADAPTATION TO MOBILE SOURCE SO2 AND
SULFATE EMISSIONS MEASUREMENT OF STATIONARY SOURCE
MANUAL METHODS
R.L. Bradow
In recent months a variety of qualification experiments on the
validity of isopropanol-water trapping of 803 with mobile source
1 2
emissions have been reported with widely varient results. '
A group at Chrysler Corporation has reported essentially unqualified
success with EPA method 83 in this application. 1 Ethyl Corporation
workers report erraneous of high sulfate values with doped SO2 both
with and without auto exhaust added. General Motors has suggested
that use of technical grade isopropyl alcohol at least partially
avoids the problem of S02 trappings in the sulfate bubbler. Ford
raises a number of issues regarding the validity of the method on
a theoretical basis^ relying mainly on supporting data furnished
by Walden Corporation in stationary source qualification experiments.
It is the purpose of this study to examine the procedures used by the
differing research groups in detail and to offer explanations of the
variable results obtained.
In the Chrysler work experiments with single cylinder ASTM
grade isooctane , engines operated on an essentially sulfur-free fuel,
were performod in order to establish the basic validity of IPA-H2O
trapping of sulfates! No apparent sulfate was found with the base iso-
-------
octane indicating that positive interferences were not present with
that exhaust. S02 tank gas in the exhaust pipe, in the sampling in
impingers or in a bag added to the iso-octane exhaust also did not
give any apparent sulfate. Lab bench studies with tank SC>2 confirmed
no SC>2 trapping or oxidation. Exhaust from 0.1% sulfur gasoline did
produce apparent sulfate and this was confirmed by gravimetric
determination of the Barium precipitate. The results of the gravimetric
method were 75% of those obtained by Sitration to a thorin end point.
Interferences from exhaust system particles and incomplete 804 trapping
were also checked experimentally and ruled out.
Ethyl corporation also operated a single cylinder engine on
isooctane with tank 502 added to the exhaust and found that 10-141 of
the S02 was retained in the first bubbler and erroneously measured
2
as sulfate. Bench experiments also indicated retention of SC>2
measured as sulfate in the first bubbler. This could be partially
eliminated by pre-acidifying the bubblers, with sulfuric acid. However,
since the added sulfuric acid constitutes a high blank, the detection
of trace quantities of sulfate involves determination of a small diff-
erence between two large quantities. Ethyl workers also reported
great variability in engine test results rendering the method
almost unusable.
The Ethyl and Chrysler experiments, therefore, are apparently in
direct conflict. Since the G. M. experiments indicated differing
-------
results with different grades of isopropanol, it is probable that
details of the methodology may account for the observed variations
in qualification test results. Therefore, the methods employed by
Ethyl and Chrysler were examined more closely.
Chrysler Corporation workers used a flowmeter followed by
series of 4 midget impingers containing 15 ml. of 80% isopropyl
alcohol (IPA) - water solution backed by a DuPont model 411 UV
photometric SO- analyzer. Sample was pulled through the train at a
flow rate of 5 liters/min. for 20 to 30 minutes to obtain a sample
of 85 to 120 liters of gas. After sampling the contents,of all four
bubblers was rinsed into a lOOcc volunetrics flask with 80% IPA
and a 25cc. aliquot was immediately titrated with 0.01 N BaCl2
solution to a thorin-endpoint. A stainless steel exhaust proble was
used to obtain the sample. Rcpeatable results were found in a wide
variety of experiments.
The Ethyl Corporation group used a standard EPA sampling train
using 100 ml. of 80% IPA in the first Greenberg-Smith impinger, this
solution plus 2% peroxide in the second and third impingers and a
fourth dry impinger followed by flow monitoring appartus. Flowrates
of 1.5 to 2.5 liters/minute for periods of 1 hour. In many of the
experiments an initial quantity of sulfuric acid of 5 to 10 ml. was
added to the first bubbler to suppress SC>2 solubility which otherwise
1 8
obscurrcd the results. Such a practice has been recommended elsewhere. '
-------
After volunetric transfer an aliquot of each of the impinger contents
was titrated with 0.0/N BaC104 to a thorin end-point. Standing time
for the impinger samples was not controlled and varied from 24 to
48 hours. The solutions were also evaporated to increase the
sulfate concentrations and, thus, the sensitivity of the method.
Cle Clearly, both groups have drastically modified the Federal
Register method in both details of glass ware, sampling and analytical
determination. In order to resolve the source of the apparent
differences an experimental program was instituted to investigate the
influence of some of the variances on the analytical results.
The experiments performed had as there goals:
1. Establishment of test repeatability with engine exhaust.
2. Investigation of possible interference in sulfate
determinations from trapped SO2-
3. Establishment of the influence of analysis variations,
glass ware type, flow rates and trapping solution variation
on test results.
-------
Method 8 sampling trains v;ere attached to the exhaust pipe
of a 350 CID 1972 Chevrolet engine equipped v/ith an autcr:at?c
transmission and operated on an engine dynanoneter test
stand. The engine v/as operated at 30 aiph and 32.0 ft-lbs
of torque, corresponding to somewhat greater than road load.
Fuels used were the EPA reference fuel with 125 ppa sulfur,
that fuel doped to 0.100 v/t. % sulfur v/ith thiophcne, and
ASTH grade isooctane containing less than 1 ppnj sulfur.
The sampling trains v/ero attached by means of Sv/agelok
fittings and a metal ball joint to a £ " stainless steel
tube welded into the exhaust pipe. Inside the exhaust pipe
the tube nade a 90° bend and a 2 " straight section v/as
faced upstream and centered in the pipe. Figure 1 indicates
the location of the bubbler trains.
Bubbler trains v/ere assembled and operated in a manner
as nearly identical as possible v/ith the trains used by
Chrysler and Ethyl . In addition, a third micro method 8
train, using ball-joint equipped midget inpingors v/as set up
in a manner recommended by the stationary source group. This
train included a single IPA-v/atcr bubbler, one -UOp bubbler,
and a dry bubbler.
Methods of analysis included BaC10; titration to a
mixed thorin-r.ethylcne blue end point for the EPA modified
train as per the Federal Register . For the Chrysler and
Ethyl • modifications v/orkup and analysis procedures those
1 P
reported in the literature for those methods wore used » .
In several experiments v/ith isooctane as a fuel a
500 ppn SC>2 in 1?2 tan1; gas v/ac; injected upstream of the bub-
bler trains through a second upstream-facing probe welded
in the exhaust pipe. Injection ratoc v/crc controlled to
-------
give a final exhaust concentration of 10.0 ppn S02. The point
of S02 injection is also shown in Figure 1.
RESULTS AND DISCUSSION;
Table 1 summarizes the results obtained with the EPA method
using 127 and 1000 ppa wt./wt. sulfur in the EPA reference fuel
and the non-catalyst engine test stand. The overall recovery
of S02 v/as good, averaging 97»5 % of the charge. The repeata-
bility of the method v/as also good, with a standard deviation
of \.7 % for 6 runs.
In all the runs there v/as apparent sulfate ranging from 9
to 37 % of the total SO.,. In experiments with added S02, ap-
parent sulfate was again detected at about 25 % of the charged
SOp. In the last three runs an attempt v/as made to purge
trapped SOp from the first bubbler by drawing through room air
at the sample rate for 10 minutes. V/olden researchers had
previously established that this procedure is helpful in avoid-
ing artifact sulfate analysis .with stack samples. However,
in these experiments, purge air had no significant effect in
reducing apparent sulfate. Thus, it appears that this pro-
cedure is capable of reproducible results but produces a sulfate
artifact of 15-25 % of the charged S02.
Samples of the first bubbler contents from runs 13 and M+
were diluted 20:1 with distilled v/ater and analyzed by the V/est-
Gaeke method for sulfite. The purpose of these experiments v/as
to determine v/hether the trapped material in those bubblers
v/ould be dissolved S02. Insignificant amounts of S02 were
found. Samples of the barium precipitate were centrifuged out
and collected for X-ray diffraction analysis. The X-ray dif-
fraction pattern shows conclusively that the material is
barium sulfate, not sulfite. Therefore, it appears that SOp
analyzed by this method is subject to oxidation of part of the
9
-------
SOg to bu^ in tne sampling process.
Table 2 presents results from the Chrysler procedure. Again,
highly reproducible results were obtained and mean and standard
deviation for sulfur recovery were 92.5/5 and 2.9# respectively.
Apparent sulfate v.ras again found, this t-imfi at. the 1/i«» level.
In this method four bubblers in series were used at rather
high flow rates for the small size of the train. Apparent sul-
fate in only slightly decreasing amounts was found in each of
the sulfate bubblers prior to the hydrogen peroxide SOg trap.
Expressed as a percentage of the apparent sulfate collected,
bubbler 1 contained 31.0#,' bubbler 2 2i+.6%, bubbler 3 22.5?$,
and bubbler 4 21 ,-6?o for the average of four runs. Tn the Chry-
sler experiments the contents of all four bubblers were combined
prior to titration and no information on the relative collection
efficiency of t.he two bubblers for real sulfate aerosols was
presented . On the basis of these tests, either the bubbler
»
train collects artifact sulfate or it is highly inefficient.
However, Ethyl data suggests that the first bubbler is fairly
p
efficient for collection of synthetic sulfate mists . In
" •
those experiments a collection efficiency of about 8Cfo was ob-
tained. It, therefore, appears that artifact sulfate is being
formed in each IPA bubbler in the train. It is interesting to
note that the first two bubblers do contain somewhat more sul-
fate than the last two. If the content of the last two bubblers
is taken to be the artifact sulfate content of the first two,
an approximate value for the real sulfate can be obtained by
difference. Thus, the real sulfate would be about 10/o of the
apparent method 8 sulfate or about 2% of the SC^. • This value
is in good agreement with sulfate values found in non-catalyst
cars by air filtration methods.
10
-------
Table 2 also presents attempts to reproduce a Chrysler qualifi-
cation experiment in which S02 was added to an auto exhaust from
isooctane which contained no sulfur. According to Chrysler reports,
no apparent sulfate was found. However, in the present study,
apparent sulfate was found in approximately the same proportion as that
obtained from the sulfur-bearing fuel. Clearly this indicates arti-
fact formation occurs with the Chrysler procedure.
Table 3 presents data obtained with the Ethyl method. Data
2
scatter was poor as Ethyl had previously found. Again, apparent
sulfate was found in all experiments including those with Isooctane.
In all three procedures, blank experiments were run with the
collecting solutions and with isooctane exhaust to insure that inter-
ferences or contaminants were not biasing results.
CONCLUSION;
Method 8 variants clearly produce apparent sulfate when used in
auto exhaust applications. It appears that this occurs by oxidation
of the S02 to sulfate, probably in the collection solution. Further
experiments are in progress to elucidate this point and to compare
Method 8, Goksoyr-Ross, and filtration procedures with catalyst-equipped
engines.
It appears that previous Method 8 work on non-catalyst engines
must be seriously in error.
11
-------
Table 1
•EPA Method
rj Fuel S Sample flow Time
10 ppn
No
10 0-100
M 0.100 |.b5
\i o.oian v.42.
O.Ci'V! a-40
ItT*
AO
£10
uo
bO
to
|V\o\es
injected - isooctane fuel
a,1
a, 40
30
30
9.01
OS?
30 .T
Cc
-------
Table 2
Chrysler Method
A: EPA Reference Fuel - 1000 ppm S
No. V^. fhcA«.«i So^~ SOa. Corwr, ^^s. * ~
I
^ 144
3 \U"4- •• u'a.-a.r a-
10.0 n
-------
Tatele 3
Ethyl Method
Sampling conditions: 3 "Lpm - 1 hour - 7.23*fmoles of sample gas
oA
Kc. 'Wrr> 'SXV r SO-j. •^O^1" SQ'i "' (teccv«.r*<\
S&.4 4R-4 3a,l4 »o.os
i.a-aR M^
looo Bl.^5
1000 5O.1 M3T 10.3 "
*•
Isooctane runs - 10 ppm S02 injected
S
-------
1. Chrysler Corporation, Response to a request for information on sulfate
emissions in the Federal Register, 39, No. 47, dated May 6, 1974.
2. Ethyl Corporation, Response to a request for information on sulfates
emissions in the Federal Register, 39, No. 47, Dated May 3, 1974.
3. A. Federal Register, 37, No. 221, Part II^ 87,075-24 and Appendix I,
Nov. 15, 1972.
B. Seidman, B.I., Ana.L. , Chem., 30, 1680(1968)
4. General Motors Corporation, Response to a request for information
on sulfates emissions in the Federal Register, 39, No. 41, dated
May 7, 1974.
5. Ford Motor company. Response to a request for information on
sulfates emissions in the Federal Register, 39, No. 47, dated
May 6, 1974.
6. Driscoll, J., ejt. al., EPA Report No. EPA-R2-72-105, November, 1972.
7. Fielder, R.S., and Morgan, C.H., Analyt. Chinica Acta,23, 538 (1960)
8. Schmidt, M., Report on Word Health Organization Project # U.S.3100,
July, 1970.
15
-------
9. Goksyr, H., and Ross, K., J. Inst. Fuel, 35^ 177(197-|2)
10. Lisle, E.S., and Sensenbaugh, J.D., Combustion, 36, 12 (1965)
16
-------
Appendix B3.5
SUJ.FATE M3TTOD COMPARISON STUDY
C2C AP2AC FPOJZCT CAPI-8-7^
R. L. Bradow
A number of researchers active in the catalyst sul-
fates field are participating in a study of methods of
analysis of sulfate in filter samples. ETCS has prepared
128 filter samples in groups of four, using a 1975 Ford
catalyst prototype and a 1972 Chevrolet which has been con-
tinuously operated on lead-free fuel. The Ford was operated
on low aulfur, 500 pms, and 2000 pns gasolines to obtain
three levels of sulfate. Low sulfate samples were obtained,
using the Chevrolet on the low sulfur fuel only. Eight
«
Highway Fuel Economy tests were run with each condition
giving a total of 32 test runs. Filters were then circu-
lated in groups of twelve to each of 8 participating labora-
tories for analysis. 32 samples were analyzed by EPA by
the barium chloranilate method,' and 10 of these were also
analyzed by X-ray fluorescence spectroecopy. Other labs
use thorin titration, barium sulfate gravimetry, HpS-methyl-
ene blue methods, and a sulfarazo III indicator method.
About half the data is now in, and the balance is ex-
pected within-the week. Some time will be required for
statistical analysis, but barium titration methods seem to
give slightly lower results than chloranilate and the Y.^S-
methylene blue method slightly higher results. X-ray
fluorescence proceeds correlate well with sample sulfate
loading up to about JfOO ug- and then seein to roll off
somewhat. At loadings of 150&u9 X-ray methods appear to
give about 30/o low results based on a linear standardiza-
tion.
17
-------
It is expected that this project will have an additional
role to pVay in conparison of S00 nethods and other methods
of test for catalyst-related emissions.
18
-------
Appendix B3.6
Determination of Soluble Sulfates in -CVS
Diluted Exhausts: An Automated Method
The initial report that catalytic converters originally designed
to reduce hydrocarbon, carbon monoxide, and oxides of nitrogen emissions
from late model automobiles also promote conversion of SO- to SO. or
H-SO, mist prompted a crash program to find or develop a fast and
sensitive methodology for sulfates applicable to car exhausts.
Although a number of analytical procedures for sulfates are described
in the literature, only a few of these have the sensitivity sufficient
to detect soluble sulfates in auto exhaust samples conveniently
collectible within the time frame of the Federal Test Procedure.
The automated method described in this report is addressed
primarily to the determination of water-soluble sulfates in CVS
diluted exhausts from cars run on nonleaded fuels. The method is
quite general, however, and may be used for trace analysis of sample
sulfates which can be leached out with water or aqueous alcoholic
solutions.
The method, first developed elsewhere (1), is based on the
reaction of sulfate ions with the solid barium salt of chloranilic acid
(2,5 dichloro-3,6-dihydroxy-p-benzoquinone). The reaction precipitates
out BaSO, and releases highly uv absorbing acid chloranilate ions, the
absorbance of which can be measured with a suitable spectrophotometer
and related to sulfate concentration. The sensitivity of the method is
19
-------
greatly enhanced by conducting the reaction in a medium less polar than
water, such as ethanol-water or isopropanol-water mixtures, where the
solubilities of both BaSO, and barium chloranilate are reduced. The
barium chloranilate method is estimated to have a limiting sensitivity
for SO, to concentration levels of 0.06 yg/ml (2).
Cations are known to interfere negatively by reacting with the
acid chloranilate to form insoluble salts. This interference is easily
removed by passing the sample through a column of cation exchange resin
in the hydrogen form. Anions such as Cl , Br , F , and PO, interfere
by precipitating out as barium salts with subsequent release of acid
chloranilate ions. Some buffer systems are reported to minimize these
anion interferences (3,5). For exhaust samples from cars run on
nonleaded fuel, ionic interference was observed to be negligible when
filtration on Teflon filters was used as a sample collection technique.
Sampling and Sample Preparation
Sampling methodology involved dilution of the auto exhaust with
air in a dilution tunnel. At the temperature the tunnel is operated,
SO reacts readily with the available moisture in the exhaust to form
H-SO, mist. The acid aerosols are sampled through isokinetic probes
and collected on 47 mm diameter 1 p pore size Fluoropore* filters at
flow rates of 28.3 liters per minute. The filters are extracted with
10 ml of 60/40 isopropanol/H.O solution (60% IPA) in capped polyethylene
*Registerd trade mark. Obtainable from Millipore Corporation.
20
-------
bottles. Extraction is accomplished by shaking the filters in ;:hc
capped bottles for at least one minute using a vortex test tube mixer
followed by a 10-15 minute soak. The supernatant extract can be
analyzed directly in the automated sulfate instrument without further
treatment.
The Automated Sulfate Instrument
A schematic of the principal components of the automated set-up
is shown in Figure 1. Hardware requirements include:
a. Reservoir (LR) for the solvent mobile phase (60% IPA).
b. High pressure liquid pump (LP) capable of delivering liquids
at flow rates of up to 3 ml/min at pressures as high as 1000
psi. Most liquid pumps used in high pressure liquid chroma-
tography would be satisfactory.
c. Flow or pressure controller (FC).
d. Six-port high pressure switching valve (SV) equipped with
interchangeable external loop (L).
e. Ultraviolet detector (D) equipped with appropriate filters
or monochromator to isolate a narrow band of radiation
centered at 310 nm.
f. Recorder to monitor detector response.
g. Automatic sampler (AS), such as the one used in a Technicon
AutoAnalyzer set-up.
21
-------
h. Peristaltic pump (PP), such as a Technlcon proportioning
pump, to draw sample into the sampling loop.
i. Cation exchange resin column (CX) - standard 1/4" O.D. x 10"
gas chromatographic stainless steel column packed with
analytical grade Dowex 50W-X2 (100-200 mesh) cation exchange
resin in the hydrogen form.
J. Barium chloranilate column (BC) - standard 1/4" O.D. x 5"
gas chromotographic stainless steel column packed with barium
chloranilate suitable for sulfate analysis.
The operating principle of the automated instrument may be
briefly described as follows:
Solvent mobile phase (60% IPA) in reservoir (LR) is continuously
fed through cation exchange (CX) and barium chloranilate (BC) columns
at flow rates of about 3 ml/min by a high pressure liquid pump (LP).
Background absorbance is continuously measured by a UV detector (D)
at 310 nm and visually monitored in a strip chart recorder. A solenoid
actuated, air operated switching valve (SV) is used for filling the
external sampling loop (L) with samples in conjunction with an
automatic sampler (AS) and peristaltic pump (PP) and injecting the
samples into the columns. At CX cations are removed and at BC color
reaction takes place. The BaSO. precipitate is retained in BC while
the acid chloranilate is carried by the mobile phase through the detector
system for colorlmetric measurement.
22
-------
For an automated sampling system such as shown in Figure 1, both
SV and PF are electrically coupled to AS and controlled by electric
timer relays such that both are activated whenever AS is sampling
(i.e. L is being filled and mobile phase bypasses L). At the end of
the sampling cycle, FP and AS stop and SV switches to the injection
mode (i.e. mobile phase passes through L and carries the sample
through CX and BC columns).
For manual operation, SV may be retained or replaced by a
similar switching valve equipped with an extended handle for manual
switching. Samples may be introduced into the sampling loop by syringe
injection or by peristaltic pump system similar to the one used in the
automated version.
The automatic sampler (AS) used in our system is a Technicon
AutoAnalyzer sampler with turntable capacity of 40 sample cuvettes.
The cam programmer was replaced by two digital timers to allow
flexibility in setting cycle times for the sampling-rinse operations.
Analytical Operation
Before the start of an analytical run, all components are switched
to the operating mode, and SV, AS, and PP are allowed to cycle normally
to clean out all components. During this time the sampling probe is
immersed in a large reservoir of 60% IPA to prevent introduction of
air into the system. Analysis of the samples can proceed once a stable
background absorbance is obtained. Sample cuvettes are filled with
sample extracts and blank solutions (60% IPA) and then covered with
23
-------
thin polyethylene film to prevent evaporation losses. The filled
cuvettes are arranged in the turntable according to the pattern blank,
blank, sample, blank, blank for concentrated samples and blank, sample,
blank for dilute samples. Blanks are used to wash out the system
between samples and minimize sample overlap. Depending on the size
of the sampling loop and the mobile phase flow rate, cycle time can
vary from 2.5 to 6 minutes per sample or blank.
Calculation
A series of sulfuric acid standards in 60% IPA is normally run
in the same manner as the samples, and a calibration curve, peak
height vs. concentration, is plotted. Sample sulfate concentrations
are calculated from the calibration curve. Total soluble sulfates
in the filter [SO~]F are calculated using the relation:
[SOpF -
-------
is richer in water. To minimize this effect, both the extracting
solvent and the mobile phase for the analytical runs should be £aken
from the same stock solution.
In order to determine the maximum absorbance of the acid
chloranilate ions as they elute out of the barium chloranilate
column of the automated system, the colored eluates corresponding to
sulfate concentrations in the range 0-30 yg/ml were collected and
scanned in a Gary 14 spectrophotometer. In this concentration range,
peak maximum was observed at 312 nm. This almost coincides with the
310 nm isobestic point (absorbance independent of pH) reported by
Schafer (3).
For isopropanol-water system, the volume of the mixture is not
equal to the sum of original 'volumes of the individual components.
In the case of a 60/40 isopropanol/water mixture, 'volume shrinkage on
mixing is about 2.7%, This 'volume change should be taken into account
when preparing standards or samples from aqueous solutions.
The working concentration range and sensitivity of the automated
system depend on sample size. A degraded sensitivity better than 0.5
yg SO? per ml in 60% IPA. was easily obtained using a 0.5 ml external
sampling loop in conjunction with a duPont liquid chromatograph UV
detector. Figure 2 shows a calibration run in the range 0-5 yg SO,/ml
using a 0,5 ml sampling loop with detector sensitivity set at 0.02
absorbance units full scale. The last two peaks, 4048 and 4048,
correspond to exhaust samples from a noncatalyst car. Testing mode was
25
-------
the Federal Test Procedure. The calibration curve is non linear with
concentration and becomes flatter at the low concentration end. This
is strongly suggestive of interplay of thennodynamic and kinetic
effects. Similar behavior was likewise observed at the high concen-
tration end.
Table I shows the precision obtained for five repetitive scans
of sulfate standards at concentrations of 1, 2, and A yg/ml using a
0.5 ml sampling loop. At this concentration range the standard
deviation is ±.05 yg/ml.
Two experiments were conducted to determine the extractability
of sulfuric acid from and absorption in Fluoropore filters. In the
first of these, known amounts of sulfuric aicd in 60% IPA were
deposited on the filters and allowed to dry overnight. The filters
were then extracted with 60% IFA and the extract analyzed for sulfates
after the filters equilibrated with the solution overnight. The second
involved immersion of dry filters in standard solutions of sulfuric
acid and analysis of the solution after overnight equilibration. The
results show that extraction is quantitative and that the filter has
practically no affinity for the solute. These results are summarized
in Tables II and III.
Table IV shows the efficacy of the collection technique for
trapping sulfuric acid aerosols. The aerosols were generated using a
Collison aerosol generator, and then fed into the CVS dilution tunnel
under conditions simulating a test run. The aerosols were collected
through isokinetlc probes and collected on Fluoropore filters. The
26
-------
back-up glass fiber filters used in these runs did not gain measu ;.»le
weights, indicating no significant breakthrough of the coliecteu
particulate from the primary collecting filters.
Figure 3 shows a typical analytical scan of extracts from exhaust
samples from cars run on nonleaded fuel. The first five pe?.ks are
sample peaks, while the next six are calibration peaks corresponding
to concentration range 0-6 vg SO,/ml. The last three samples were
diluted tenfold to bring detector response within calibration range.
As a general rule, calibration runs are always made for each series
of samples, as peak height-concentration relation may change as flow
rate, back pressure, and column permeability vary over an extended
period. This practice may be dispensed with for systems equipped
with integrators.
Table V shows typical results of analysis for soluble sulfates
of nonleaded exhaust samples collected on Fluoropore filters using
the Federal Test Procedure. The low sulfate results correspond to
test runs with noncatalyst cars and the high results to test runs with
catalyst equipped cars.
A few filter samples were analyzed sequentially by x-ray
fluorescence technique and by the barium chloranilate method. The
filters were first analyzed x-ray fluorescence, then extracted with
60% IPA and analyzed for sulfate in the automated instrument. The
results are summarized in Table VI. Considering the fact that sample
handling techniques were not closely monitored, agreement between the
two methods is e:^ our aging.
27
-------
Conclusion
The automated method described in this report offers a sensitive
(less than 0.5 yg SO~ per ml), fast (less than four minutes throughput
time from initial sample injection into the column), and convenient
method for the analysis of soluble sulfates in auto exhaust. Sample
preparation is minimal, as this involves only simple extraction with
60% IPA. There are no precipitates to cause deterioration of the
optical cell, as the BaSO precipitate is effectively retained in
the barium chloranilate reactor column. Although primarily addressed
to trace sulfate analysis of auto exhausts from cars run on nonleaded
fuels, the method may be adapted to any sulfate sample which can be
leached out with water or aqueous alcoholic solution.
28
-------
Table I
Precision of Repetitive Measurements
Peak Height
[SO ] in yg/ml
9.7 21.2 47.8
9.9 20.4 48.8
9.6 21.2 49.5
10.2 20.3 48.6
8.8 21.2 49.0
Mean 9.6 20.9 48.7
Standard Deviation ±0.5 ±0.5 ±0.6
Coefficient of 5.2 2.4 1.2
Variation
29
-------
Table II
Recovery of Deposited H SO on Fluoropore
Filters by Extraction with 60% IPA
Total pgs SO. on Filter
4
Deposited Found
10 10
20 20.5
30 30
40 40.5
50 50
60 60
169 172
338 350
507 494
30
-------
Table III
Absorption of H SO in 60% IPA $y Fluoropore Filters
Total ygs SO in Solution
Initial Final
10 10.5
20 20
40 40.8
60 61.2
200 205
400 392
31
-------
Table IV
Collection of Generated H.SO Aerosols
Fed into the CVS Dilution Tunnel
Sample #
4001-3
4002-4
4003-2
4004-1
4005-3
4006-3
4007-1
4008-2
Mass Loading
in ygs
956
1791
1076
1323
2403
296
468
21181
Total SO. on
Filter in ygs
350
664
390
217
856
115
197
8438
% SO. on
4
Filter
36.6
37.1
36.2
16.4
35.6
38.8
42.2
39.8
32
-------
Table V
Typical Results of Sulfate Analysis of Nonleaded
Exhaust Samples Collected on 'F"luoropore Filters
Mass Loading
Sample # in ygs
4034-1 415
4035-3 271
4036-3 252
4037-3 151
4038-3 120
4039-3 287
4076-3 232
4079-3 308
4080-3 430
4084-3 506
4087-3 765
Total SO
in ygs
20
15.5
16.7
11
10.8
10.5
84
106
192
241
316
% SO as %
Mass Loading
4.8
5.6
6.6
7.3
9.0
3.3
36.2
34.4
44.6
47.6
41.3
33
-------
Table VI
Soluble Sulfate Analysis: Preliminary Comparison of
X-Ray Fluorescence and Barium Chloranilate Method (BCM)
Total SO on Filters in yg
Sample #
4006
4007
4014
4017
4023
4032
4036
4038
4039
4050
Mass
Loading
459
379
358
285
390
1065
224
84
250
390
X-Ray Fluorescence
Low Resolution High Resolution
208
184
143
37
142
296
12.8
17.0
12.4
18.0
BCM
219
173
156
44
113
245
9.8
7.8
9.8
13.7
Ratio
X-Ray/BCM
0.950
1.064
.917
.841
1.256
1.208
1.306
2.179
1.265
1.314
-------
References
1. Bertolacini, R. J. and Barney, J. F'., "Colorimetric Determination
of Sulfate with Barium Chloranilate," Anal. Chem. 29, 281 (1957).
2. Ibid, "Ultraviolet Spectrophotometric Determination of Sulfate,
Chloride, and Fluoride with Chloranilic Acid," Anal. Chem. 30,
202 (1958).
3. Schafer, H. N. S.7 "An Improved Spectrophotometric Method for the
Determination of Sulfate with Barium Chloranilate as Applied to
Coal Ash and Related Materials," Anal. Chem. 39, 1719 (1967).
4. Barton, S. C. and McAdie, H. G., "An Automated Instrument for
Monitoring Ambient H SO Aerosol," In Proceedings of the Third
International Clean Air Congress, Dusseldorf, Federal Republic of
Germany, 1973, VDI-Verlag Gmb H, 1973, p. C25.
5. Gales, M. E., Jr., Kaylor, W. H. and Longbottom, J. E., "Determination
of Sulphate by Automatic Colorimetric Analysis," Analyst 93, 97 (1968).
35
-------
FIGURE 1
FLOW SCHEMATIC FOR AUTOMATED SULFATE INSTRUMENT
LR
RECORDER
*»-TO WASTE
TO WASTE
-------
Figure 1
Plow Schematic for Automated Sulfate Instrument
LR - Liquid reservoir
LP - High pressure liquid pump
FC - Flow or pressure controller
P - Pressure monitor
SV - High pressure switching valve
L - External sampling loop
CX - Cation exchange resin column
BC - Barium chloranilate column
D - UV detector
FM - Flow monitor
AS - Automatic sampler
PP - Peristaltic pump
37
-------
5.0
0.002 ABSOR!
ANCE UNITS
4.0
0.5
j>4_~jj T^v—^
oo
ro
4049
-------
Figure 2
Sulfate calibration for concentration range 0-5 \ig SO
per ml in 60% IPA. 4048 and 4049 are exhaust samples
from a car not equipped with catalyst.
39
-------
1/1
QJ
1
00
* 1
o ^
II ^J-
-o cr
-------
ENGINE ROOM DILUTION TUBE FLOW CHARACTERISTICS
BY
Robert L. Holden
Emissions Testing and Characterization Section
Chemistry and Physics Laboratory
National Environmental Research Center
Research Triangle Park, North Carolina
National Environmental Research Center
Office of Research and Monitoring
U.S. Environmental Protection Agency
Research Triangle Park, North Carolina 27711
September, 1974
-------
ABSTRACT
This report describes work done to characterize gas flow and
particulate flow in a dilution tunnel used in automotive emissions
research. In addition to velocity profiles being graphically
presented, a particulate profile s^udy is also included. A descrip-
tion of some troubles and corrective measures is also reported.
Finally, some qualitative statements, based on rough preliminary
measurements, are reported on the subject of aerosol deposition and
loss to the walls and bends of the engine exhaust pipe of our system.
-------
ACKNOWLEDGMENTS
The author greatly appreciates the help and support of all his
section coworkers. But special thanks is given to Dr. Ronald L. Bradow
for suggesting the importance of this study and to Henry J. Becker whose
help in fabricating necessary parts and carrying out the measurements
was invaluable.
-------
I. INTRODUCTION
A series of experiments was conducted on the 18 inch diameter
engine room dilution tube located at the Beaunit facility of EPA-NERC,
Chemistry and Physics Laboratory. These experiments were designed to
determine the velocity profile and aerosol particulate concentration
profile for this dilution tube. In addition, some qualitative data on
sulfate aerosol deposition and loss was obtained.
II. VELOCITY PROFILE
The purpose of these experiments was to determine if the velocity
distribution of the tube is conducive to proportional sampling among the
four filters of the probe rake which is placed at the down stream end of
the tube (See Figure 1).
The measurements were made using a Thermo Systems Inc.* model 1054B
anemometer and model 1210-60 hot film sensor. The air flow was produced
by the tube exhaust blower. The sensor was placed approximately two
inches in front of the sampling probes and was moved horizontally and
vertically along lines passing through the tube center. Data was collected
at one inch intervals, with no data taken closer than one inch from the
tube edge (to prevent damage to the sensor). After the initial profile
run, it was determined that some changes were in order and 2 more profiles
were made.
-------
cn
LEFT
BOUNDARY
118 in. DIAM.I
[ABSOLUTE
i FILTER i
EXHAUST
PIPE
1 ORIFICE MIXINGl
{PLATE I
(LEFT BOUNDARY!
ro CRITICAL I
FLOWVENTURl!
IGHT BOUNDARY!
|FOURlln.DIAMJ
I PROBES I
(Side view of 18-Inch dilution tunnel|
-------
The results of each profile are shown on graphs of horizontal
distribution and vertical distribution which jFollow this discussion (See
Figures 2, 3, 4, §5). Note that test "A" shows higher flow velocity in
the top left quadrant of the tube, and in general a large fluctuation
across the tube (turbulence was also rather high for this test). This
'was deemed unsatisfactory and investigation revealed two possible problem
areas:
(1) incomplete filling of air bag filter
(2) large open spaces (located in upper left quadrant) between tube
boundary and flow control baffle.
Test "B" was conducted with the air bag filter removed and the
baffle sealed to the tube boundary by silicone sealer. This produced a
relatively smooth profile across the tube (turbulence still present, but
reduced somewhat).
Test "C" was performed with a absolute filter of negligible pressure
drop in place of the air bag filter (baffle remains sealed). This again
produced a relatively flat (about 5% fluctuation across tube) profile
with reduced turbulence. We believe this velocity profile to be
satisfactory for aerosol data collection by filtration methods.
-------
15.0
g
:<
o
•5
4.0
M
•LU
B
(3.01 L
I9j 18
TOP OF
TUBE
UJ Ijy
IJj 111
lit
j DISTANCE FROM CENTERLINE, inches [ .
[Figure 2. Vertical velocity distribution in 18-inch gas dilution tube, test 'A'.]
5.0 t
4.0
itn^
I BOTTOM OF
I TUBE
|5.0
4.0
3.0
Hi
LEFT BOUNDARY OF TUBE
(FACING GAS FLOW)
i DISTANCt FROM CENTERLINE, inches '
iy
[RIGHT
5.0
4.0'
1
a
3.0 i
BOUNC'.RYOFT.BEl
Figure 3. Horizontal velocity distribution in 18-inch gas dilution tube, test '*'
-------
e
ITEST c -ABSOLUTE FILTER;!
BAFFLE SEALED
B • NO FILTER;
BAFFLE SEALED
m
| DISTANCE'FROM CENTERLINE, inches.
Figure 4. Vertical velocity distribution in 18-inch gas dilution tube, test 'B' and 'C'.
n
5.0
«fc
3.0
LEFT BOUNDARY OF TUBE
i (FACING GAS FLOW)
DISTANCE FROM CENTER LINE, inches [_
5.0
TEST C - ABSOLUTE FILTER;
BAFFLE SEALED
TEST B-NO FILTER;/
BAFFLE SEALED
I Figure 5, Horizontal velocity distribution in 18-inch gas dilution tube, tests 'B' and *C*.
RIGHT BOUNDARY OF TUBE ,
FACING GAS FLOW
-------
III. AEROSOL FLOW PROFILE
The purpose of this group of experiments is to examine tho mass
concentration profile of aerosols carried down the tube by air fl from
the exhaust blower unit. The information obtained from these e ^jr :ncnrs
allows us to determine if the aerosol follows the gas flow uniformly
or if it collects in certain areas of the tube cross section .
A model 7300 aerosol generator from Environmental Research Corporation
was used to generate a sulfuric acid mist from a 10% solution of the acid.
The aerosol flow rate was 49 liters per minute which corresponds to a mass
generation rate of approximately 1.04 grams per minute. The mist was
then injected into the dilution tube at the point where the engine exhaust
pipe opens in the tube. As in the velocity profile tests, the dilution
tube exhaust blower created an air flow which carried the aerosol through
the tube. Note that the engine was not operating during any of these tests,
The aerosol was sampled by a one inch diameter stainless steel probe
drawn across the tube horizontally and vertically (passing through the
center). Figures 6 and 7 illustrate the sampling points. The aerosol
was drawn into the sampling probes by a vacuum flow of one cfm, and was
then collected on teflon fluoropore filters of 47 mm diameter. Each
sampling run lasted 30 minutes.
The results of the experiments are shown in figures 6 and 7. This
data indicates that the aerosol mass concentration profile is relatively
smooth and uniform (at least to the limits of reproducible massing
49
-------
!1100
l/l
o
TOP OF
TUBE
ly [2j 111
I DISTANCE FROM CENTERLINE. inches i
Figure 6. Vertical aerosol profile in 18-Inch gas dilution tube. |
lit'
UJ IT
I BOTTOM OF
log]
TUBE I
I',.
LEFT BOUNDARY
OF TUBE
(FACING GAS FLOW)
! 6
!J_! |JJ I 2
DISTANCE FROM CENTERLINE, inches
LL! LU
Figure 7. Horizontal aerosol profile in 18-inch gas dilution tube. |
II! I 91~
RIGHT BOUNDARY
OF TUBE
(FACING GAS FLOW)
A V
-------
accuracy with a microbalance). An independent chemical massing
performed on the filters confirms this uniformity, and we are convinced
that the aerosol follows the gas flow and will be sampled representatively
by the four-probe sampling rake at the end of the tube.
To examine the possibility of aerosol loss due to engine exhaust pipe
interception of the mist, the aerosol was injected just inside the exhaust
pipe so that it had to travel through four feet of 2 inch diameter pipe
with one right angle bend. The aerosol was sampled by the four-probe
rake (which is in the center 4 inches of the tube) and collected on 47 mm
fluoropore filters. The results (shown in Table 1 below) indicate,
qualitatively at least, that there could be a 10% to 20% aerosol loss to
the exhaust pipe walls and bends. Thus, before sulfate studies are
carried out, one should test this loss possibility and make corrections
to data if necessary.
TABLE 1
PROBE f
1
2
3
4
DIRECT INJECTION:
AVERAGE MASS COLLECTED
(y grams)
718
483
958
518
INJECTION INTO EXHAUST
PIPE: AVERAGE MASS
COLLECTED
(y grams)
592
422
764
468
NOTES: (1) averages taken over three runs.
51
-------
'AN EPA AUTOMOBILE EMISSIONS LABORATORY
RESEARCH TRIANGLE PARK, NORTH CAROLINA
by
Peter A. Gabele
Ronald L. Bradow
Chemistry and Physics Laboratory
Research Triangle Park, North Carolina
52
-------
ABSTRACT
This report describes the establishment of an exhaust emissions
laboratory where gaseous emissions can be analyzed in accordance
with the 1975 Federal Test Procedure (FTP) and particulate
emissions are conditioned for sampling in dilution tunnels. The
effect of fuel additives, catalysts, and other pertainent variables
on the character of both gaseous and particulate emissions are
examined in this laboratory.
The paper discusses the measures taken on the test stand to
attain simulation of a vehicular engine when operating over a
specified driving cycle. Test equipment, experimental apparatus, data
analyses methods, data reducing methods, and calibration and evaluation
measures of significance are described. The results of a comprehensive
aerosol study on the 18 inch diameter dilution tunnel are provided
which include velocity and concentration profiles and information on
particle deposition within the tunnel.
53
-------
SECTION I
INTRODUCTION
PURPOSE
Concern has been leveled at the extent and character of participate
emissions from internal combustion engines. Much of the work
accomplished to date has dealt primarily with the effect of lead anti-
(1 2)
knock and lead scavenger additives on exhaust emissions. ' More
recently emphasis has been placed on characterizing particulate emissions
from engines which are equipped with catalysts. The effect of fuel
additives upon these emissions is also in need of investigation.
Therefore, project personnel have endeavored to establish an engine-
dynamometer test facility having a particulate sampling capability.
The particulate sampling function does not interfere in any way with
the ability to sample exhaust gases in accordance with the 1975 Federal
Test Procedure.
Particulate matter exhausted from automobiles is defined as any
material, other than unbound water, which condenses at 90°F into
particles larger than a small molecule, but smaller than 500 microns
in diameter. These emissions are of concern because particulate
matter suspended in air has been designated by EPA as a criteria pollutant
for which ambient air quality standards are required. Compliance with
these standards mandates the development of emission control regulations
which eventually may apply to automobiles.
-------
SCOPE
In order to compare project findings with results from chassis
dynamometer studies, test stand engines must closely simulate actual
vehicular engine operation. This requirement specifies such things as
test engine type, engine parameter settings, test cycles, and mechanisms £01
properly loading the engine. In short, loads and speeds of the test
stand engine must follow as closely as possible those of that engine
when operated in a vehicle over an identical test cycle. The extent
to which simulation is being carried out should be determined once
the set up becomes operational.
Design and development of a particulate sampling apparatus is
required. Representative samples must be collected in a form which
can be readily analyzed for composition and weight. The sampling
system must be evaluated to determine the extent to which a represen-
tative sample is being collected.
The flexibility of being able to examine different engines, test
cycles, catalysts, and fuels and fuel additives is necessary. Operation
of the Constant Volume Samplers at various dilution ratios is also
necessary.
All system variables of ordinary interest should be either
continuously monitored or easily obtainable. For this purpose a
comprehensive temperature measuring, indicating, and recording system
is necessary. Also, such things as engine and dynamometer torque,
sample flowrates, etc., should be displayed on gauge boards or console
indicators.
Two independent gasoline engine-dynamometer systems are required
55"
-------
to provide sufficient experimental data consistent with project goals.
Both engines should be capable of simultaneous, continuous, and automatic
operation. This requirement' is imperative for mileage accumulation pur-
poses.
A data reduction system must be developed to reduce emissions
information. This system must be easily accessible and capable of
yielding rapid results. Emissions of hydrocarbons, carbon monoxide,
and nitrogen oxides must be reduced to units of grams per mile in accordance
with the 1975 Federal Test Procedure.
56
-------
PROJECT OUTLINE
DESIGN AND CONSTRUCTION PHASE
The design and construction phases of the test facility
the installation and, in some cases, design and calibration of the
following:
1. an engine dynamometer bedplate,
2. two eddy current dynamometers,
3. two engines,
4. two dilution tunnels,
5. two constant volume sampler (CVS) systems,
6. particulate sample probe rakes and filter holders,
7. engine transmission to dynamometer couplings,
8. lube oil and fresh water cooling systems,
9. flywheel inertia weights,
10. temperature indicating and logging system,
11. programmable cycle driver.
Replacement of various outdated dynamometer indicating and control
instrumentation was also accomplished in the design work.
Design of the dilution tunnels and associated sampling equipment
is of vital importance to the validity of the experiments at hand. Figure
1 shows a schematic of the dilution tunnel geometry for one test engine
set up; Fig. 2 shows the other tunnel. For ease in future identification
of these two systems, the 4 inch and the 18 inch diameter tunnels will
hereafter be referred to in connection with X system and Y system,
respectively.
57
-------
I > • i i ' '. <
,
: i , | .
; l ; i
'1:1
' ! i ;
, ' ' t !
f '
I
!.. i
;.!'!.;
l- I '
.1 i.. ;
'. ' . j • i l i i : i ; j
i ' .! i I ! j. i
• '. i i j 1 M ! : ' • I I', i
• : : i"' '•" [ < "i '" 1 i !" > "•
: . , •. : i'I i., ; . i i j'
'-: •:!!•; iiiH;;::-1 ';;
i ' i I ' ' i i ':'| 1 ; ; i , . '•' i i :
• ! ' '. , !' ; i'1 ; i
I ! ! I : .::,,"
1 L. .U.14 LL»!_1_L! :J •_:
' :l:H.:-;ri:.|:ri '• jl
U i s ! rii r IT! i ; i •irri: -:S -J-.r
i > i
i '1
•AKOMTElj-;-
i h *"?-'!.!:
.1.1
;- ' !•-
;fH--P
•pi TtH
i3'Tul
LJ r • t
.'j "'"j' .[1 ."L'iV-i .j. FJ
:i.t'L*.Jf.fjjur.jJ,i i
-.!J.'4«^**!4-4^-"?
II*' 1 * 1 v'S' '• "4
'7~! T "Mi 7 i";~ i i ".%'.*
rciJL tx r~i''i A2: ''.+•> * '"j
ri'":-'1';TJ
~~i " "i "*T" ~!""-r\r^"i
•J-f -, 4—4 -j -J^-r-i - • **-'
{......'". .."."...',. ....; .;.''•
|;-.::::ni|;.!:|Hi
jt^iJilli-ii-!11
i_i. LL.i.'i i j j r^* j ' - T •* ' H~| "(J"r"r"-~H"iTj ^^
.''.'j4.jp.DIAMLJ:;:! i: 4^.rj\_ :K ±miiTO
• .._:,'.' r-;.; i -. :-: J..ImftM.r-* .••—.._ ^jj
! ! '
| I I ' i ! I I ' ' i I ! I'll ''•'!! I | I | I ! ! , I I i
. • ;i." '"''•'! ":";": f:' : I ' ! IT1"1 if; N'T. jVl ';"•"!": : i"n I ;"i"i"['| ~P~r RTr!' r:~t~l"~
'.! ' ' I ' I ! ' ! • j ' ! ' ; : I.' I. ' I. If I LJ
i : M ; • ; ! i ' ; • ! : i M ' ; : I I ' I • • : , . • ! ' ' •' : ! | I | f , fM I'1"'" ,"l..r
l I j i ! i I i : i ' ' , i i . i i I I I " I" i ~ i" i
i , i : ,.,;.: i j * 1
'.'•••• ; ! ' ; ' . i • : ;
• « • . ill ! • i
I. j ! • •*'" -fc '.,-.-'. _L.L_'_.:._. <_,_J.J._ '_L CZ\ ! \ IU , LJ
' i ! llNJl ! -L U^—i. ^ ^—RJ '" - '» ---r:«==S^ -^H—'-f? VACUUM H:
i '.' T- .1 IvT ! .!.!„ l.i l.jj.i I if : • .« • i.i.j • 1 i . i.i i !. . -iJ / 'i"/7sJ _h.J SOURCE i"
^ \-\'.': j \ -'.'LII ' . T i i i"/ ^/4"'i !Mr^115! 1"
•! ' ' ! ' ! i s*^ ' rl J i i i ! iFXHAiisTt ; M i i ' ;. I ! i . / ' ; V , ; 7!; i'!+
; : | ! '—^^ i i. Li ' i.i [j;h: ]TW01-in.OIAM, .147-mmFILTER,! rM fi
l| iiij i i .Fill !. Fi ii^^ ilillMVlij -b^s^:p^i:n-cj:|..:
-'.;.' • • : '' ;..; ! .j. LLi ;J_ i • M.JJ. i i . ! , '- • • r i ~Luj U
nil :i l.-ili itn, j F«.« • -IT•;• i i -\i\nT4sS~
Side view ot 4-Inch dilution tunnel. : i .i-|-|..| j- j-|- -p+EI -
| I M ' i!; J'I i-rmrr-n'j-.i'v'":'"•-:"! j iv I - - I- J•• -+-H-1-f
il!!!:!|.;i!|!ii!i:i.!|iU. I'-U il-li-i!^" ' n|ffipffl|fOTS
jiltiti^illrfiftmlfil^ f^|ffflS^p|^ra
i M I M i ! ! !'! i ' M'il I I j! j i i ! ! : i ; i s I J i -; i ri ,M- •
j , . , , t ; i i ; ; ; . , , | j . i . j , , . , . -. , i . , j i , , i ! , • l ;•( - - ) iTi T: it±i-
: • i i •-.; | • • • • : ; . i | i • i i i i !.i ' • .': i ; ! ; i i ; : | i i.. i : 14 .! 'i .! L*L|7' "Ilpluri"
M ' i i ! ' ; I ' • 'J. ''• > ' L_!..i • i :'' ' t i i i"' ' i ! I i" ' ' i : ' ; ' i ' '
a i
'M
'— I— —-*--— - r ' '-'i -.
'. ::TL|:i;"l:i±t.;5'_i
.r LI ix:±i±n.j
i ! I l i !•:•'.
.MUiltuifH
~|7~r CpXJ jl'mlTi
L LlTill 1-0.1144
II i I I I ! ' i • • •
-------
1/1
to
JRIGHT I
1 BOUND ARY-L
"AJR] a
• IN 1 m
JLEFT— 4-j
{BOUNDARY]
i
,1
\
X.
If
u
.'ABSOLUTE! EXHAUS'
'FILTER f PIPE
, . J
^K U8 in. DIAN1.I
X s*-*
ORIFICE MIXING j 6 in. DIAM.)-^ s^z
PLATE 1 ACCESS I (
^ /
1 FOUR 1 in. DIAM.
i PROBES
1
n
^
^ ^
\ ^^
1 plLEFT BOUNDARY1
Li7±jjTO CRITICAL |
-fnM FLOW VENTURI
T 1 — LPIRUT nnuwnflov
J |6-in. DIAM.
^V^ —
fcL_|lTO VACUUM
^ f SOURCE '
1 Side view of 18-inch dilution tunnell
-------
EVALUATION PHASE
Test stand engine operation and the dilution tunnel required
evaluation for qualification purposes. The following evaluation
procedures were conducted:
1. The test stand engine speed-manifold vacuum traces and exhaust
gas emissions were compared with those of the vehicle on the
chassis dynamometer.
2. Gas velocity and particle concentration profiles were examined
inside the 18 inch dilution tunnel.
3. The extent of particle deposition was examined in both dilution
tunnels.
CONSTRUCTION AND DESIGN PHASE
General Layout
The plan drawing in Fig. 3 shows two principal areas wherein the
project is concentrated—the engine-dynamometer room and the analysis-
control room.
The engine-dynamometer area contains two gasoline engine-dynamometer
combinations, two dilution tunnels, two constant volume samplers, and,
eventually, a diesel test stand and particulate sampler.
To achieve good sampling results, both temperature and humidity
control are necessary. Ten tons of air conditioning is available to
the project area for this purpose. Temperature in the engine-dynamometer
room are controlled to about 72°F and humidities are maintained below
65%.
The analysis control room contains three engine dynamometer consoles,
two cycle programmers, a NO-NOX chemiluminescent analyzer, one NDIR C02
and two NDIR CO analyzers, one FID hydrocarbon analyzer, and an
60
-------
CYCLE PROGRAMMER,:
Y SYSTEM I
er>
DYNAMOMETER CONTROL,
IY SYSTEM
I DYNAMOMETER CONTROL,
/i X SYSTEM
OOO(OQ
| ZERO AND SPAN GAS |
CYLINDERS
r
!_YJ
JCONSOLEJ
| CONSOLEJ
/
DIESEL I
CONSOLE:
GAS ANALYSIS
| EQUIPMENT!
o o o o
GASOLINE ENGINE
TEST BED
J MACHINE t
SHOP |
DIESEL ENGINE
TEST BED
\-\& 3; Plan vi- •/ of project area.
-------
automatically sequencing three bag sampling system. Windows
between the control and engine rooms permit closed door operation
with all personnel stationed in the control area.
Engine-Dynamometer Stands
Initial construction began with the installation of a 20 ton,
5 ft. by 20 ft. steel bedplate. The bedplate rests on a 10-inch thick
reinforced concrete floor. Holes have been drilled into the concrete
and anchor bolts are used to secure the bedplate in place. Isopads
between the bedplate and floor are arranged to effectively isolate
any bedplate vibrations generated during engine operation. The engines
and dynamometers are oriented on the bedplate as shown in Fig. 4.
Both engines may be operated independently, however, when run simultan-
eously, gas emissions from only one engine can be sampled at any one
time.
The engines are 1972 model, 350 CID, Chevrolet Impalas. Both are
equipped with 2-barrel Rochester carburetors and 350 turbohydramatic
transmissions. These engines and transmissions are identical to those
installed in the Chevrolet test vehicles at this facility. Engine oil
and water are cooled in heat exchangers which are plumbed into the
building water system. Water thermostat valves on the engines have been
retained and are set to maintain water temperature below 195°F.
Transmission oil is cooled in a fin type, air cooled, heat exchanger
with a 12 volt automobile air conditioning fan.
62
-------
X
ENGINE
DYNAMOMETER
DYNAMOMETER
Y
ENGINE
BEDPLATE
F'G ^/l Bedplate configuration.
63
-------
A 55 gallon drum located outside the building for safety reasons
supplies fuel to the engines. Fuel additives can be added, mixed,
and set into place on the supply line within minutes. This arrangement,
however, has precluded the use of fuel evaporative cannisters.
The engine-dynamometers are dry gap, eddy current types
manufactured by Eaton Dynamatic. Both are absorption rated at 175 hp.
Dynamometer cooling water temperature, rpm, current excitation level,
and torque are displayed on console guages located in the control room.
The dynamometer controls apply loads to the engine by controlling
excitation current to the dynamometer in accordance with two different
modes:
1. Speed Control: The excitation current is varied as necessary
to hold the engine at a selected speed providing, of course, the
throttle is sufficiently opened.
2. Current Control: The excitation current is held constant while
the speed is allowed to vary with throttle position.
These modes may be selected either manually or automatically
during operation. For example, the dynamometer controls are programmed
to automatically switch to the speed control mode when braking the
engine during periods of rapid deceleration.
Engine Control Features
Each engine has been equipped with a throttle control actuator
which is part of the closed loop servo system designed to control engine
throttle from a program source. The engine-throttle-actuator system
is set up to control the servo loop around speed by utilizing a rate and
-------
value feedback from the engine. On the Y engine an additional rate
feedback is obtained from a manifold vacuum transducer for stabilization
purposes.
Each engine is inertially loaded by utilizing a flywheel which is
keyed to the dynamometer shaft. The total inertia load on the engine
is calculated as a composite of the dynamometer rotor, couplings, and
flywheel inertia weights. This is 306 and 508 units for X and Y systems,
respectively. The required inertia leading based on a vehicle weight
of 4500 Ib. is 793.4 units. This inertia deficit has been compensated
for in the Y engine system by programming loads into the dynamometer
during periods of acceleration. Unfortunately, with an eddy current
dynamometer there is no way of putting energy back into the engine,
therefore, coast-down decelerations with the installed system tend to
be higher than normal. This tendency is illustrated in Figure -T
wherein a comparison between coast-downs of the vehicle and X engine
are compared. Although the disparity between these results is obvious,
from an emission standpoint the error induced because of this is probably
negligible when operating over the EPA Urban Driving Schedule.
Engine Calibrations
Various steps were performed in order to simulate actual vehicular
operation. Carburetors were adjusted to attain the proper air-fuel
mixtures by measuring exhaust gas content until it reached 0.5% CO and
2.5% ©2 levels (slightly lean of stoichiometric during idle) . The
programable inertia load was adjusted until a wide open throttle(WOT)
acceleration from 0 to 50 mph in 8.0 seconds was recorded. This
corresponded with the WOT acceleration on the vehicle. Excitation to
65
-------
O1
01
SIMULATED, SPEED, mph ;
]<*. 5" I Coastdown time for vehicle and X test-stand engine. I
-------
the engine dynamometers was adjusted such that the 50 mph level ro '1
load corresponded with that of the vehicular engine on the chassis
dynamometer. Manifold vacuums were used in matching engine load-.
The excitation necessary to match level road loads at speeds other
than 50 mph approximated closely the 9 percent level required at 50 mph
(see Fig. G). Therefore, an excitation of 9 per cent is set into the
dynamometer during steady cruise periods when the dynamometer is in
current control mode.
Dilution Tunnels and Constant Volume Samplers
Both engines exhaust into dilution tunnels. The dilution tunnels
and their associated sampling equipment are fabricated from stainless
steel. The X system tunnel was designed by Esso Research and Engineering
(4)
Company to measure vehicle particulate emissions. The Y system
tunnel, although designed by project personnel, is similar to a 16 inch
tunnel which was developed by The Dow Chemical Company.
For sampling particulate matter, the Y system tunnel has a four
probe rake and the X system tunnel has a two probe rake. All probes are
made from a one inch diameter stainless steel tubing. Each probe is
connected to a vacuum source through a 47mm filter holder, a flowmeter,
and a regulating valve. Flowrates are maintained at 1 ACFM throughout
a test.
The 4 inch diameter tunnel (X system) is connected to a CVS which
utilizes a roots blower as the constant volume source. The 18 inch
diameter tunnel (Y system) utilizes a critical flow venturi as its
constant volume source. Flowrates in both systems (see Figs. T and 3 )
67
-------
CT>
CO
1 *»l
i
Is s
JEXCITAT
! o,
— ij
1
^^*"
1
BJ IJOJ
i
,'<=> £ 1 Current excitation
1 1 j_ U—
•^•^^^^^^•^^^^^^
"
1 1 1 1
|J5J |4Q | | 451 | 50) [55J
i SIMULATED SPEED, mph i
required to simulate level road load of vehicular engine.!
-------
0)
(O
i4-in. DIAMETER DILUTION TUNNEL!
TWO-PROBE
PARTICULATE
SAMPLE
RAKE
! TEMP. AND PRESS.
SENSOR
CONSTANT VOLUME!
SAMPLE TO GAS
ANALYSIS EQUIPMENT
'BLOWER I
MOTOR I
.'ROOTS I (PRESSURE GAGE'.
; BLOWER!
«*7 X system dilution tunnel and CVS arrangement.]
-------
FM T°FLRTE| 1 "-in. DIAMETER DILUTION? !
— -y5 — J TUNNEL ! (
\
TEMPERA!
SENSOR
\
K
i 'EXHAUST
{ PIPE
fUREj
1
1
^^
FOUR-PROBE
PARTICULAT
SAMPLE RAKI
CONSTANT VOLUME
.AMPLE TO
SAS ANALYSIS
EQUIPMENT^
^
^
i
E
i
•— «
^
I CRITICAL FLOW
1 VENTURI |
/
.
, Y system dilution tunnel and CVS arrangement.
•TEMPERATURE Ap
PRESSURE!SENSG
CENTRIFUGAL)
BLOWER I
-------
can be varied by changing either drive belt pulley diameters in the X
system or venturi sizes in the Y system. Once the system is set, total
gas flowrate through the tunnels is determined by inducing a known
quantity of propane into the system over a given period of time. The
concentrations of propane in dilution air are determined using FID
analyzer on a bag sample. Actual flowrate, Q, is calculated by the
formula :
Q = (M) (22.4) (35.315) (IP"3)
(t) (c/106) (44.1)
M = mass in grams of propane injected
t = time in seconds over which test is carried out
c = concentration of propane in ppm
constants = 22.4 I/mole
35.315 ft3
44.1 g/mole C3 H
ID'3 m3/l
Flowrates for the X system and Y system are 342 and 432 ACFM,
respectively. Since the Y system employs a critical flow venturi as its
method of maintaining a constant flowrate, the actual flowrate will vary
slightly with pressure and temperature changes according to the relation-
ship:
Q = CP// T
P = pressure at the venturi entrance
T = temperature at the venturi entrance
C = constant
Because pressure changes in the installed system do not vary
significantly during operation, only temperature changes require monitoring
in order to correctly determine actual venturi flowrates. This function
is fulfilled by temperature indicating system. Temperatures at the
inlet to the roots blower on the X system CVS are likewise monitored. In
71
-------
both systems the temperatures are used to correct actual flowrates to
standard values when calculating exhaust emissions.
Emission Analysis and Reduction
Both gaseous and particulate emission information is sought.
Gaseous emissions are reported in accordance with the 1975 FTP.
In addition, levels of fifty-five (55) individual hydrocarbons are
measured using gas chromatographic techniques almost identical to
those reported by Dimitriades and Seizinger . Energy dispersive
x-ray fluorescence spectroscopy is used to perform trace metal
analysis on particulate matter. Particulates are also catalogued
using a scanning electron microscope. SO- is determined by an
(8)
adaptation of the method of West and Gaeke and sulfate analysis
are conducted using an automated colorimetric procedure involving
the reaction of sulfate ions with the solid barium salt of chloranilic
acid.(9)
Presently, data reduction for gaseous emissions is accomplished
through a PDP-12 computer. Programs to compute emissions of hydrocarbons,
CO , and NOX in grams per mile, and fuel economy in miles per gallon,
are stored on magnetic tape. Two of these programs in FOCAL-12
language are listed in Appendix A.
EVALUATION PHASE
Evaluation of the engine dynamometer set up was conducted to determine
how closely the engines were simulating actual engine operation in a
vehicle. An experimental program was also developed to evaluate the
velocity and particle concentration profile in the plane of the sample
probe inlets.
72
-------
Engine manifold vacuum traces were recorded over the EPA Urban
Driving Schedule (LA 4-5-3). The records for both test stand engines
and the vehicular engine are shown in Figs. 9-J/. X system engine
displays consistently higher manifold vacuums when compared with the
other two engines. Comparison between the Y system engine and the
vehicular engine is good and indicated that both of these engines
are experiencing approximately equal loads. Prior to recording all
traces, engine manifold vacuums were observed equal under steady state
50 mph cruise conditions. Therefore, the differences seen in the traces
are due primarily to the differences in inertia loadings. X system
engine's high overall manifold vacuum is easily explained by citing its
undersized flywheel for which compensation has not yet been made.
Other expected differences appear when comparing exhaust emissions
reported over the EPA Urban Driving Schedule. Emissions are shown in
Figure 12. The X system engine exhausts consistently lower NOX
emissions because it is experiencing lower combustion temperatures.
Again the cause is the lower inertia loading experienced by the engine
during acceleration.
Both velocity and particle concentration profiles were determined
in the probe plane of the 18 inch dilution tunnel. Actual determination
of profiles for the 4 inch tunnel is physically complicated by the
small tunnel diameter, however, a strong theoretical argument supporting
the contention that uniform profiles exist in the cross section of the
(4)
tunnel has been developed by Esso Research. Although sample flowrates
at both tunnels was maintained at 1 CFM; only in the Y system can this
be said to approximate an isokinetic condition for obvious reasons.
73"
-------
' I
MM i-i-U-.j-.u ;.' i" U '
> . • • i I * ! < I '1 ll
-t— ••;-j •• ;•-' : •«-'
i i ! !»1*1
ij~:'"'"; ! ";.'i ' i" i
11 .J.L!i: LLLJ.L . ij' i rl'
-Li ! ' ' U !. LJ.-U.Ll J_IJ. i ' : :.
! i
iii ; i i i i
! ! I I : " 1 I 1 S '"' !
! t±t -;.
< i ! -: ' i ! I
; I ! ! i i i
'!)
o |Manifold vacuum trace for X test stand engine from the 160 to 280 sec, mark on
; i i i i ; ' ' ' ^: l ) " • ' : i l i ( ; r ! ' : i i i
i ,
LA-4-5-4 cycle.
-------
I. I.'.! 1 j • ' Tl '.L"U :"_' i
.,-^i • .-<-..•-!.'
J-4.4Ai.LLj.jIi.
. L.'—i -i J J r. |..L
Ul
- i !;;rn ,
'" "'
:..• fTIME, seconds! : • i.i. !
-
.. ._ .. .. • * . _ . a...o.-i.
.iciManifold vacuum trace taken for Y test-stand engine from the 150 sec, to- 280 sec.. mark. on LA-4-5-4.cycliJT
— — -
' |. i I..:: .[...]..•. i ,j ;• ,L.'. !
I I ' i ' I i I i T J i i I , ,
l|i j : i'i 111 I i j | > '
i ] I • ; Li ! ! I ! ! ' J
I . ' ' ; _|_! i i J ! jj..1.' .1
1 I ! I H i.j'l i
i • • •• : •
-------
04 .
225P ' J.'j"
r i I .
II
[215)
JTIME, seconds'
Figured. Manifold vacuum trace taken for vehicular engine from 150 sec, to 280 sec. mark on LA-4-S-4 cycle."- ' i
-------
4
1 t
1J-.
.
2
o
s
UJ
L_2j
i
2*1
z
!_!_;
!oi
1 1 1
a
o §"
a
o
o
o
o
o
o
.
•
o
o
1 1 1
1 W '
UoJ
,
f
UJ
a
X
0
i
E
CO
a:
g
20|
Oil
1 1 1
o
— o B-
o a
a
o a
• 0
• o
-1
— —
1 1 1
LU
Lii
I
c/»*
o
CO
oc.
s
o
a:
a
^.
l\
•Ul
1 1 1
— —
— —
O
* ffi 1
l-
1 1 1
IX j |Yj 'VEHICLE' |X| 1 Y; (VEHICLE f X> |Y \VEHICLE
iTEST-STANDf TEST-STAND) TEST-STAND I
i ENGINES | ENGINES / ENGINES *
I ENGINE IDENTIFICATION
Vic\. ;i. JGomparison of emissions for x and y test-stand and vehicular engines.!
-------
Because exhaust gas particles are generally below 3 microns,
their behavior is much like that of a gas. For this reason the
importance of sampling isokinetically becomes somewhat academic.
Velocity profiles were determined both horizontally and vertically
while traversing with a thin film anemometer. Initially, traverses
were performed with bag-type filters upstream of the tunnel. Profiles
constructed for these runs indicated a non-uniform velocity situation
and, therefore, absolute filters were substituted in their place.
Traverses with and without the absolute filter upstream of the
dilution tunnel resulted in the rather uniform profiles shown in Fig.^3j
Concentration profiles were established using a sample probe and
fluoropore filter to sample particles generated by a Collison particle
generator. A 10 percent solution of sulfuric acid was dispersed by
the generator as an aerosol having a particle size
range from 0.03 to 3 microns. A 1 inch sample probe was traversed
and particles sampled through the probe were collected on a teflon
fluoropore filter. Each traverse point was sampled for 30 minutes
and the sample flowrate was held at 1 CFM. Following collection, the
filter weight was determined both by weighing and by wet chemical
analysis of H^SO. content on the filters. Both determinations agreed
reasonably well. The horizontal and vertical concentration profiles
are shown in Figs.15- /&, . The profiles indicate that particles are
unrt.
well mixed evenly distributed in the plane of the probe rake.
78
-------
0» ,. ft
.«« :4.U
O
3 in
i uJ 1 3.0
12.0
— '1
BOU
1 (H
1 1 1
^^
\ \ \
Oj :9| '_8i ;_6|
LEFT
YDARY OF TUNNEL
ICING AIR FLOW) !
1 1
-^xr — v.
i i
1 1 1
| ABSOLUTE FILTER !
~+»ff^"
| NO FILTER,
1 1 1
L*J lit i! 'I1 ll! li-
^DISTANCE FROM CENTERLINET inches
1 1 ^
•*•*.„-**«•
— -**v.
— 4.0J
— 3£j
I 1
)8, ,9 115.-^
T
I RIGHT
JBOUNL- °v OF TUNNEL
1 (FAC ' AIR FI/'W)
Velocity profile on Y system dilution tunnel during horizontal traverse '
-------
00
o
.5.0
'5
IP-
i i
!1
4.0
'M
20
-^
«r •.
—
~—
1
"10 .91 8'
1 UPPER'
BOUNDARY OF f UNNEL
1 (FACING AIR FLOW)
1
^•^-^••••••" *x
^^^^
\ \
"****V-
I4v^
1
1
1 1
CA>
(NO FILTER f
««^
^ J ABSOLUTE FILTER I ^*-^
1
1
>6i 4j -21 !_!_' il 111 li'
1 DISTANCE FROM CEBITERLINE, inches!
Fie^l Velocity profile on Y system dilution tunnel during •horrzortta* traverse./
1
(1; 19
BOUNDARY
(FAC1NGAI
—
~
\ u i
4.0 ^»
3*1^
p:
O
^
_ n tij
3.U >»
2.0!
f 10
OF TUNNEL!
R FLO») \
-------
.!.';« l i
rh.LI ;.'
t I
I .J '.-< '-
--M -i-- i 4-|-
: IJ r-j . ; i.
Ili-H:!!!
-L J_..UJ.L .{--U^-jXi ! I ;
! • ' ' < I l.| .h i.i.l I I '...-: Li- -.
i I L'a-jVLLl jJ. :.h-U! '.J
! T. I I I' : I ' . i .I.J4.I_1..:
••••-
! T ! '"'! • J LLlj
: i i i i i ' —
i iiT i11' il'
i i I..J '.i l.:_i L
. i i i • i
Lj.i.L
. . '.i-i -u-i- '
1 ! ! I.LLlj.
I i i l r'~1
101 | 191 i |8J
- ;T''M
LEFT
BOUNDARY OF TUNNEL
=-l '
DISTANCE FROM ICENTERLINE, inches!
I
__.!JBOUNDARY OF TUNNEL:
(FACING AIR FLOW) \\ '^Filter weight plotted as a function of probe location during horizontal traverse.^ (FACING AIR FLOW) [
' j " ( ! " i >
r Till"1
U-t;
I '• I
I • 1
I.L
i 1
I '
I
i I.I •! ;i
'
I : : i
i :
i i
I '
i i
j
! I
T.S r. -l-tj ; i ! n
I I ! • ' I !
ri^iii
• I
! __ J L.L ' 1-1-: .-' • H -*" •
i I ill LlJTi L! I
V ' | i >'! LLL ... • ' ! L
I i
j :•
,T
i '• 1.11.!.! Ll..!.L:.U. -!..;:-.'.
" •' ' . ! ! I i i I . ' . • I i
i I : i ' • M L..L I
\.\ ' ' i i ! i i i.• ' ' •
} i . [ i i ti j i i - . . i '
i ! ! l.b ! I i ! * !.! !
. i . i i i i i i I ' i i i ; !
i M I | M i i.i I i |j i
! i ' i i I I I i iJ !_! ' i '
V; ; ! [" ' '? i" i~ "' ' ' '
I
i 1
i ITIJt'i r:i:i i M ^T I".! :.:
•J'r r: "it I , ; t i i l •
i 1 {- I' !•
"i"4 ~i r' f • i
_l_l.J.:'_i_LLL' LJ '. L':_LLD-;- —
i ! LlTi"
r i i r i
I ! ,
i i i
• i
ITT
\
-------
1 i
• i
•:;:i:;:HMj;MHijU;,H;r.:±!
ji i;-! jil-Ml- -;,!-:: i-;. -H-iT-j it
':i; :.i': tp: '.rnrhr
...... ji :_LpL||.|
ILrtr. I [.a:j H.
. '
! '
1200
.UJ
'800
I i I i I t '
1 ' » ' ' 1 i ' '
';' , ; rj'iJ~M •" , i i '."
. •' • i . ; i ' i i~ i
I j .*. :
t i j
IQT-U
Itt; i r i
±Qjj .'..!
L
i H-H-h-H
4J__
TH ,
* 4J}..L.B-.Li
< i I I ! i
1 T
UPPER
.BOUNDARY
.OF TUNNEL
.: . ! L !. . i..!.L. Ml
L|_ I '-LiJ'U. Ji ..L 14.
i I i Li i; r M i. ;n r
. I.;.; I'LL .lL_"7]J ..'
<. i DISTANCE FROM CENTERLINE, inches!' _L' ! •
f-r -p ... I.'..!•• rr|.| | |
! • ! j.ilSH i M iU>lj ri \ li
M : ; ! ; i u"i"i LM i :~.<
. to
Mi°
-. |iop
I I
L.J 'j L^J LOWER!
BOUNDARY : :
lOFTUNNELi
F»'e\.l4.(Fliter weight plotted as a function of probe location during verticaLtraverse.l "i • '
>j i ,. i •"-- - -T i .— . ' ill i i 'i i • i —i
l ' • M • i | • ' : I | ! I ' i | | , , • \>
! : . • • - i l • ' • l • i , ! ' • !. i , i" I : ' i
i i "l < .( !"i MT M '« I
I ! i . M IT": \ >"*'\ 'I
I i :
-------
Particle deposition studies were performed on the 4 inch
dilution tunnel, but not performed on the Y system, 18 inch tunnel
because such studies had been previously conducted on an identical tunnel,
In both cases the results were determined as follows1:
1. The engine was operated over a specified number of cold
start LA-4-S-3 cycles.
2. Particulate samples were taken and the total particulate
emission was calculated.
3. The tunnels were washed with dichloromethane.
4. The wash was evaporated and the residue weighed.
5. Percent deposition was determined.
The results had indicated a 3 percent deposition of particulate in
CL
the 18 inch tunnel as compared with 5.9 percent value in the
4 inch tunnel.
The 5.9 percent level in the 4 inch tunnel compares with a value
(4)
of less than 1 percent found by Esso. Since actual engine exhaust
was used in the EPA study as opposed to the monodesperse aerosol
(3.5 micron diameter) generated in the Esso study, the smaller,
warmer exhaust particles were probably subjected more by the influence
of thermophoretic forces resulting ultimately in a greater deposition
upon the cold tunnel surfaces.
83
-------
SECTION II
CONCLUSIONS
1. Reasonably good simulation of actua"1 vehicular engine operation
has been achieved in the case of the Y system engine-dynamometer set up.
2. The X system does not adequately simulate vehicular engine operation
but it does yield repeatable results.
3. The 18 inch dilution tunnel is capable of representatively sampling
particulate emissions from gasoline engines.
4. The emission of hydrocarbons, carbon monoxide, carbon dioxide, and
nitrogen oxide can be accurately determined from both X and Y
systems in accordance with the 1975 Federal Test Procedure.
-------
SECTION III
RECOMMENDATIONS
1. One of the reference test vehicles should be equipped with
a drive shaft torque meter. This would enable the measurement and
recording of instantaneous values of torque and engine rpm versus
time, hence, accurate speed-load curves could be generated for com-
parison with those on the engine-dynamometer test stand.
2. In order to reasonably simulate vehicular engine operation, the
X system engine-dynamometer set-up should be inertia compensated. A
means should be implemented for inertially loading the engine such
that speed-load curves correspond with those of the vehicular engine
when driven over identical cycles.
3. The 4 inch diameter dilution tunnel should be operated on the Y
system engine to compare particle collection data with the 18 inch
tunnel. Such experiments would hopefully indicate the effects, if
any, of sampling non-isokinetically.
4. Instrumentation should be installed to enable modal analysis
of both gaseous and particulate emissions. A better understanding
of the relationship between engine mode of operation and the resulting
character of pollutant emissions is needed to better predict
atmospheric quality as a consequence of emission control measures.
85
-------
SECTION IV
REFERENCES
1. Environmental Protection Agency, Federal Register, Vol. 37, No. 175,
Title 40, Part 85, Sept. 8, 1972.
2. J.B. Moran and 0. J. Manary, Interim Report, PB196783, "Effect of
Fuel Additives on the Chemical and Physical Characteristics of
Particle Emissions in Automotive Exhaust", NAPCA, July, 1970.
3. K. Habibi, "Automotive Particulate Emissions and Their Control",
SAE Paper 710638, October 24, 1970, Midland, Michigan.
4. M. Beltzer, R.J., Campion, and W.L. Petersen, "Measurement of Vehicle
Particulate Emissions", SAE Paper 740286, Feb. 25 - Mar. 1, 1974,
Detroit, Michigan.
5. EPA Contract EHS-70-101, The Dow Chemical Company, APTD-1567,
March, 1973.
6. B. Dimitriades and D.E. Seizinger, "A Procedure for Routine Use in
Chromatographic Analysis of /Automotive Hydrocarbon Emissions",
Environmental Science and Technology, Vol. 5, No. 3, p. 222,
March, 1971.
7. T.G. Dzubay and R.K. Stevens, "Ambient Air Analysis with Dichotomous
Sampler and X-Ray Fluorescence Spectrometer", Paper submitted to
Environmental Science and Technology, May, 1974.
8. K. Klostermand and R.L. Bradow, "Direct Determination of Sulfur
Dioxide from CVS -diluted AutomExhaust", NIEHS Symposiu, op.cit.
9. S. Tejada, "Determination of Soluble Sulfates in CVS Diluted
Exhaust: An Automated Method", NIEHS Symposiu, op. cit.
86
-------
APPENDIX A - Program Listings
Programs for reducing data are written using FOCAL-12 language on a
PDP-12 computer. Two of the programs, one for reducing gaseous
emission data from Y engine and the other for computing fuel
economy on the X engine, are listed below.
1. Program to reduce gaseous emission data obtained during an EPA
Urban Driving Schedule:
L '-» /KM( 1, 1
A LI
«; i-OCAL-K-i
JJ1.!)L> A "If** IM DMA" A,3,C,l).£,K>..bJ,AA,(,,La,fci
'H.O'l A 'J.l)>'.Jo,S>.i',l}a,0;j,n,b.i,ii,i,jf |
o l • 3 «- b '\i = - y
01. DV .-, L'.i=v.n^*..jHi).)i633; b CU=b-TKl-l/
0 -OJ b iJ]j=v,..-.:lliiJ..)5/Mf,fv'=Cij: b hfc.= i,-**( 1
3^ J7*Kh; b .NJ = \H-1
™.'w s uiu??i ,%ri"""S M-"I!>
(IfeJ^ft I X.-y/M'Sv^t ! : 1 7.,"»|."f/f •
ite. 73 (i 01 ./i 9
03. ni r =.
87
-------
2. Program to calculate fuel economy in miles per gallon:
C rOCAF.-ltf
Ol.Oci -\ "1V-=K I v) rJCOM L'^M" At B* C, \), t. HA, (-, U. .^, 0, h, J-, I , XX
01. 4 b b V=( 17.3b.i33J) "«( K) *( . 3ti.-J37rfb) MC-UO fSr'i/'/i .J"-Vi
i)tJ) ) ; i> v/= 1 / ( 1 - . r)-.)4 V it ( H- 7 b)
01. bJ b •{= 13./I/C f-+Cifi v
01. bb b < = ;-.'-i^( l-l/f)
01. b7 b :3D»V/*5f:7. K . il:)c)i) 1 b33 : b CC- b-T * ( 1 - 1 /
ill . 63 b l)|J=V./fe. u.JOjD£»16
Ul.6'1 b /r=(t-<1<>(l-l/^
Ol.e-b .^ ^7.= ^,, jc. i)fjjJ3«i J7
i.lrJ.(S3 b »'i>=n/i^3/C (.','!*
Oa.^b 1 ". ."MOf p,«i", •'.],!; f /'."CU-I pvj"t MM, !
OJ.66 f •:,"v...", v,, ! ; i ,',,MUr", f , !
•.)y.f:7 I ?, "MO >;-rF*M, OP, ! : T ,1, "IHC lf, PH, !
?i')3.a-{ 0 16.37
The variables listed in both programs are identified as follows:
A = relative humidity, %
B = saturated vapor pressure of water, mm. of Hg.
C = barometric pressure, mm of Hg.
D = sample inlet pressure, mm of Hg.
E = time for bag 1, sec.
E2 = time for bag 2, sec.
E3 = time for bag 3, sec.
AA = temperature at inlet to venturi, °R
G,G2,G3 = concentration of C02 in bag 1, bag2, bag 3, respectively; ppm
U,U2,U3 = concentration of CO in bag 1, bag 2, bag 3, respectively, ppm
P - background concentration of CO, ppm
Q,Q2,Q3 = concentration of HC in bag 1, bag 2, bag 3, respectively, ppm
R = background concentration of HC, ppm.
S,S2,S3 - concentration of NOX in bag 1, bag 2, bag 3, respectively, ppm
T = background concentration of NOX, ppm
XX = background level of O, %
88
-------
Appendix B3.9
Status Report
ROAP 26AAE
Task 007
Protocol to Characterize Gaseous Emissions
as a Function of Fuel and Additive Composition - Prototype Vehicles
This task was begun as a protocol development project at the Bureau
of Mines. After completion of the basic program, it appeared that addi-
tional work was needed to respond to the non-regulated emissions milestones
of January 1, 1975. Consequently, a new contract program was begun at
Southwest Research Institute to determine fuel additive protocol test
procedures for PNA, sulfur and nitrogen compounds, and such other species
which may be of health effects interest as the in-house program proceeds.
Future work will evaluate the control system performance protocol as well
as performing tests on non-catalytic prototype engines.
Status:
Two water-brake dynamometers and control systems have been set up
and 350 CID Chevy engines have been mounted and broken in. Test
methods for active hydrocarbons, aldehydes, COS, H2S, S02, S03, phosphine,
PNA, nitrosamines, ammonia, and phenols have been set up and calibrated.
Mileage accum-lation tests on the base fuel are in progress. Two other
fuels, a high aromatics fuel, and the base fuel containing a full additive
package trace levels of TEL and high sulfur will be run. It is antici-
pated that this program will be completed by late fall.
89
-------
Appendix B3.10
Status Report
ROAP 26AAE
Task 009
Protocol to Characterize Part1'culate Emissions
as a Function of Fuel and Additive Composition
Concept:
Again the concept associated with this task was to rely on a sub-
stantial contract effort at Dow to develop a standardized test method
for particulate measurement. The rather small in-house effort was
designed to validate Dow procedures and to extend sulfate characteriza-
tion data to a variety of catalytic and non-catalytic prototype automobiles.
The Dow studies on Chevrolet vehicles and engines produced highly
variable data. A statistical analysis of the Dow in-house particulate
data is in progress in order to establish control levels for a draft protocol
in progress.
The in-house work involved careful qualification testing of particulate
sampling devices to establish their reliability with both sulfuric acid and
organic aerosols. A paper on the EPA CVS-compatible system was presented at
the NIEHS Symposium in April and a recent report on results with Exxon and
EPA systems is included within Appendix B3. A report on non-catalyst and
catalyst systems including determination of fuel sulfur level effects was
prepared for submission as SAE paper 740528. Both monolithic and pelletted
catalysts have now been studied in some detail.
90
-------
Appendix 83.]]
SOUTHWEST RESEARCH INSTITUTE
8500 CULEBRA ROAD • POST OFFICE DRAWER 28510 • SAN ANTONIO, TEXAS 78284
August 10, 1974
TO:
FROM:
SUBJECT:
PREPARED
FOR:
Dr. Ronald Bradow, Project Officer
Environmental Protection Agency
Research Triangle Park, North Carolina 27711
Charles T. Hare and Karl J. Springer
Department of Emissions Research
Southwest Research Institute
8500 Culebra Road
San Antonio, Texas 78284
Interim Report on Phase I and Monthly Progress Report
No. 13 for the period July 1 to July 31, 1974; Contract
No. 68-02-1230, "Development of a Methodology for
Determination of the Effects of Diesel Fuel and Fuel
Additives on Particulate Emissions, " SwRI Project
No. 11-3718.
Environmental Protection Agency
Research Triangle Park, North Carolina 27711
Introduction
The purpose of this project is to develop and demonstrate a test
procedure suitable for characterizing the effects of diesel fuels and fuel
additives on particulate emissions from automotive (truck and bus) diesels.
The scope of this work includes construction and use of a dilution tunnel
for diesel exhaust, use of a variety of particulate sampling devices and
techniques, and application of a wide range of chemical analyses to diesel
particulate. This report covers the test protocol which has been developed
to meet project objectives, including all assumptions and calculation tech-
niques.
Conclusions and Final Results for Phase I
The Phase I effort, by agreement with the Project Officer, included:
the development and construction of all necessary items of equipment; de-
velopment of all necessary test and data reduction procedures; development
of chemical analysis procedures for dimethylnitrosamine and phenols; and
demonstration of the entire system protocol using one engine (Detroit Diesel
SAN ANTONIO. HOUSTON, CORPUS CHRISTI, TEXAS, AND WASHINGTON, D.C
91
-------
6L-71T) and one fuel (6, or emissions test No. 2 diesel). This scope
of tasks under Phase I is that required under the contract except for
the number fuels involved, but it was agreed that demonstration on
one fuel would be sufficient to meet contract objectives.
A. Specification, Procurement, and Assembly of Equipment
The dilution tunnel itself was constructed of 3. 18mm (1/8 inch)
stainless steel sheet rolled into a tube of 273mm (17.7 in) inside dia-
meter. It was made in two sections, eacli 2.44mm (8 ft) long, with
flanged connections at the mid-point and at both ends. The tunnel is
shown in Figure 1, along with: the dilution air cleanup filters (extreme
left); the upper end of the high-volume sampling system (extreme right);
the exhaust muffler and transfer pipe (below tunnel left of center); and the
engine air intake system (vertical duct at left). The amount of exhaust
flowing into the dilution tunnel is controlled in two ways. First, in addi-
tion to the small vertical transfer tube leading from the muffler to the
tunnel, the muffler has two other (larger) outlets with gate valves to con-
trol flow through them. Closing the valves forces more sample into the
dilution tunnel, and vice versa. As a second control parameter, the
diameter of the perforated tube (inserted into the muffler) through which
exhaust must pass to enter the (nominal) 3 inch O. D. transfer tube was
varied to arrive at best maximum and minimum flowrates. The best dia-
meter was found to be (nominal) 1 1/4 inch O. D. tubing. A sketch of the
dilution tunnel is given as Appendix page A-2, including the internal de-
tails and critical dimensions.
Figure 2 is a detailed view of the adaptations made to the (nominal)
4-inch sampling system originally specified, including the tapered inlet
(reduces cross-sectional area by about 33 percent) and the transition
made to a (nominal) 8-inch by 10-inch high volume sampling system. This
system originally had a 108mm (4.25 in) diameter inlet and used 102mm
round filters. Flow through the high-volume system is set and measured
by a calibrated orifice mounted in a "tailpipe" affixed to the blower ex-
haust.
The other sampling system used on the diluted tunnel flows is a set
of four smaller units which are operated simultaneously. Figure 3 shows
the portion of this system which is inside the tunnel when in operation,
consisting of four probes mounted near the tunnel centerline. These probes
have a nominal inside diameter of 12. 7mm (0. 5 inch) at the tip, and the
stainless ducts are standard 1-inch O. D. tubing. A sketch of a probe is
included as Appendix page A-3. In operation, the probe system is mounted
as shown in Figure 4, with a 47mm stainless steel filter holder on each
"arm". Flows through the four-probe system (or "4 x 47" system) are set
and measured by the instruments shown in Figure 5. The four flowmeters
are used to maintain isokinetic sample rates, and the total flow through
92
-------
Figure 1. Diesel particulate dilution tunnel
Figure 2. Modified 4-inch sampling
system
Figure 3. 4 x 47mm probe
system
93
-------
Figure 4. 4 x 47mm probe system
mounted in tunnel
Figure 5. Flowmeters and dry gas meters
used with 4 x 47mm sampling system
Figure 6. ERG sampler
and peripheral equipment
-------
each system for a given test is measured by one of the dry gas (totalizing)
meters.
In addition to the tunnel techniques, particulate is collected during
some runs by a diluter-sampler developed under a separate EPA contract.
This "ERG sampler" is shown in Figure 6 (center), along with the sampling
pumps, flowmeters, and dry gas meter (left) which are necessary to its
operation. The particulate sample is actually collected in the 47mm filter
holder below the ERC unit, and the exhaust sample for the ERC is extracted
from the exhaust pipe just upstream of the muffler as shown in Figure 7
(connector in bend of elbow is probe insertion point). The sample probe
used for the ERC unit has a tip inside diameter of 5.09mm (0. 2005 in) and
most of the fabricated sample line is (nominal) 1/2 inch O. D. stainless
tubing. A sketch of the probe is included as Appendix page A-4. A section
of (nominal) 1/4 inch O. D. stainless tubing 152mm (6 in) long was inserted
between the 1/2 inch line and the ERC.(nominal) 5/8 inch diameter sample
line to decrease sample flow somewhat.
Two other techniques are being used to evaluate particulate emis-
sions from the engines under test, and they are both qualitative from the
particulate mass standpoint. These techniques are the Federal (PHS)
smokemeter, and the Bosch EFAW sampling/spot reading system. Figure
8 shows these techniques being applied back-to-back, with the Bosch
sampling probe inserted into the pipe and the PHS optical unit in position
on the end of the pipe. The PHS meter reads smoke plume opacity (re-
motely), and the Bosch system employs a reflective reading from a fil-
ter through which a standard volume of exhaust is drawn to place a value
on smoke intensity. The PHS meter is the standard tool used to certify
diesel engines in the United States for smoke performance, and the Bosch
unit is widely used in engineering test and evaluation.
Up to this point, the engine used for testing and developmentfwork
has been a Detroit Diesel-Allison 6L-71T unit, shown in Figure 9 as set
up for operation in the test cell. The other engine to be used in the pro-
gram is a Cummins NTC-290, to be tested after all six fuel configurations
have been tried on the Detroit Diesel. A limited amount of testing has also
been performed using two diesel-powered automobiles, an Opel and a
Nissan. Figures 10 and 11 are two views of the experimental setup used
for these light-duty vehicles, with a long exhaust pipe extension (insulated)
to reach the dilution tunnel. Any regularly-scheduled test program on
light-duty vehicles would probably utilize a tunnel mounted near the dyna-
mometer rather than the existing tunnel, since the existing one is really
intended only for bare engine usage.
The other major items of equipment needed for this test program
are a microgram-sensitivity balance and a temperature- and humidity-
controlled environment to house it. The system constructed for this
95
-------
Figure 7. Direct exhaust sampling
point for ERC sampler
Figure 8. Federal (PHS) smokemeter
optical unit and sample acquisition for
Bosch EFAW unit
Figure 9. Detroit Diesel 6L-71T
in test cell
96
-------
Figure 10. Exhaust duct used for tests
on Opel and Nissan dies el autos -
first view
Figure 110 Exhaust duct used for tests
on Opel and Nissan diesel autos -
second view
•
Figure 12. Humidity- and temperature-controlled
chamber housing microbalance used for gravimetric work
97
-------
project is shown in Figure 12, and it has yielded excellent results. The
humidity control system consists of a large insulated spray chamber/
water tank/chilling unit through which all intake air is drawn, and a reheat
coil at the chamber outlet to control temperature within ± 1°C. Absolute
humidity has been observed to vary only slightly as the chilling unit cycles,
on the order of 0. 05% water vapor in the air. All air entering the chamber
has been filtered twice by MSA Ultra-Aire systems, and the air is not recycled.
B. Selection and Procurement of Test Fuels and Additives
In order to fulfill contract objectives, it was desired to employ
a variety of fuels and additives in evaluating the test protocol. Variation
was considered necessary in fuel boiling range, sulfur content, and hydro-
carbon type composition (paraffins, olefins, aromatics). The specifica-
tions finally agreed upon are given on page B-2 of the Appendix, and the
properties of the fuels as received are given on page B-3. Fuel A (EM-
197-F) is similar to a No. 1 kerosene fuel such as might be used in a
municipal bus fleet. It has a rather low boiling range and density, about
10% aromatic hydrocarbons, and very low sulfur content. Fuel B (EM-
195-F) is essentially a No. 2 diesel emissions test fuel, with a broad
boiling range and about 35% aromatic hydrocarbons. This fuel has re-
latively high sulfur content and the highest density of the three test fuels.
Fuel C (EM-198-F) falls between fuels A and B in density, fraction of
aromatics, and sulfur content. The low end of its boiling range is trun-
cated, however, because it is a specially-blended fuel and does not con-
tain the normal range of base stocks. The three fuels were procured in
amounts of at least 11,400 1 (3000 gal) each and stored in specially-
designated tanks at the Department of Emissions Research.
The additives agreed upon for use in the project are Ethyl DII-2
and Lubrizol 8005. Ethyl DII-2 is a primary hexyl nitrate, and is used
as an ignition (or cetane) improver. Treatment level varies, but the
most commonly used percentage is 0. 1 percent by volume. Maximum
treatment level is cost-limited at about 0. 15 percent by volume, at which
point refinery methods become more economical in upgrading fuel quality.
Lubrizol 8005 is an organo-metallic, containing calcium and a
small amount of barium. The recommended dosage is 0. 25 percent by
volume, representing the optimum cost effectiveness of the material as
a smoke-suppressing additive. There is apparently some concern that
treatment levels above 0. 25 percent by volume may lead to increased
ash deposits over the long term, so the level used for this project would
certainly not exceed that value. Both additives have been procured in
amounts sufficient for project use.
C. Chemical Analysis of Particulate Samples
Several types of chemical analysis have been set up to process
98
-------
samples of diesel particulate taken during this project. To begin, a
commercial laboratory has been retained to perform quantitative S,
C, H, and N analysis on samples deposited on 47mm glass fiber filters,
as well as S, C, H, N, and O analysis on organic solubles. It is cur-
rently planned to submit 56 samples per engine for S, C, H, and N
analysis, and 20 samples of solubles per engine for S, C, H, N, and O
analysis. The total test/analysis matrix will be described later in the
report, showing exactly which modes and fuels will be analyzed.
Another commercial laboratory will be performing trace metal
analysis on samples collected on 47mm Fluoropore filters (0. 5^rm mean
flow pore size). The metals to be analyzed for are summarized in Table
1, along with nominal detection limits in/yg/cm^ filtration area and
<<(g/filter based on a nominal effective filtration area of 13 cm^. Calcium
and Zinc were included in the test array at the last moment because a
trial sample showed them to be significant. The current test plan calls
for submittal of 96 samples for trace metal analysis, which should yield
a good picture of variation in particulate metal content with fuel and
additive composition.
Analysis of particulate samples for total soluble organic content
is being performed by a research group at Southwest Foundation, the
Institute's sister organization which deals mostly with basic medical re-
search. The solubles are extracted in chloroform, and subsequent
TABLE 1. METALS TO BE ANALYZED AND
NOMINAL DETECTION LIMITS
Nominal Detection Limit
Element 4g/cn/ yg/ filter*
Ba 0.24 3.1
Ca 0.12 1.6
Mn 0.10 1.3
Pb 0.14 1.8
Sn 0.26 3.4
Ni 0.10 1.3
Cu 0.10 1.3
V 0.12 1.6
Sr 0.14 1.8
Zn not known not known
* based on effective filtration area of 13cm
analysis for BaP is performed by thin-layer chromatography. Spectral
analysis of the soluble fraction is performed by NMR and IR, and the
remaining sample is split. Tor SCHNO analysis and paraffin determination.
99
-------
Copies of NMR and IR spectra run on early practice samples are in-
cluded as Appendix C for information purposes.
Paraffins will be separated from total organic solubles by liquid
chromatograph, and a boiling point distribution will be obtained by FID
using ASTM D 2887-72T or a similar technique. This analysis will be
performed on a work order basis by the U. S. Army Fuels and Lubricants
Research Laboratory, which is located on the Institute grounds and staffed
by Institute personnel.
The SwRI Department of Chemistry and Chemical Engineering has
been working under this contract to develop analysis methods for nitro-
samines, phenols, and organic peroxides in diesel particulate. The pro-
cedures for nitrosamines (n-dimethylnitrosamine in particular) seem to
be working well on test samples, and the phenol technique appears to have
adequate sensitivity. The rectangular glass fiber filters used to collect
samples have a variable and sometimes quite significant phenol back-
ground level, however, and filter washing has been only partially successful
in eliminating the problem. A new type of high1"- "cleaned-up" filter is
currently coming on the market, and we should receive our first batch in
a few weeks. It is hoped that this development will yield significantly
better results. Tests designed to yield samples for phenol analysis are
being skipped in the current test schedule, and these specific runs will be
made later after the problems have been ironed out.
Efforts were made to find an approach to analysis for organic per-
oxides, but no real success was achieved. The level of effort allocated
for development of analytical methods and actual tests did not permit a
very broad-ranging development effort, and it was decided that the attempt
for peroxides should be dropped to avoid jeopardizing the number of actual
evaluations needed for phenols and nitrosamines. A summary of the me-
thods developed for phenols and DMNA is given in Appendix D along with
some early sample chromatograms.
D. Development of Test Plan and Data Matrix
Since the total number of samples being taken for this project is
extremely large, a test plan was devised to meet project objectives while
avoiding unnecessary duplication. The information gathered from the tests
conducted should be ample to decide which types of analysis lend them-
selves most readily to fuel and additive qualification. The test plan and
data matrix are summarized in Table 2. Following this plan (with no
extra runs) yields the following numbers of independent analytical deter-
minations on each engine;
1. 312 gravimetric (other than ERC)
2. 48 trace metals
100
-------
TABLE 2. TEST PLAN AND DATA MATRIX FOR EACH ENGINE
Fuel(s)
B
B + 8005
Operating Condition
Speed
Idle
Peak Torque
Peak Torque
Peak Torque
Peak Torque
Peak Torque
Rated
Rated
Rated
Rated
Rated
Load, %
...
0
25
50
75
100
0
25
50
75
100
Composite
Idle
Peak Torque
Peak Torque
Peak Torque
Peak Torque
Peak Torque
Rated
Rated
Rated
Rated
Rated
...
0
25
50
75
100
0
25
50
75
100
Composite
Analysis Codes* by
Sampling System
4v 47
X 1 i
1258
1258
1 5 8
1258
1 5 8
1258
1258
1 5 8
1258
1 5 8
1258
1258
1258
1258
1 5
1258
1 5
1258
1258
1 5
1258
1 5
1258
1258
Hi-Vol
First Run
1346
1
1
1346
1
1
1
1
1
1
1346
1346
1346
1
1
1346
1
1
1
1
1
1
1346
1346
Rpt.
1 7
1 7
1 7
1 7
1 7
1 7
1 7
1 7
* Analysis Codes
1. Gravimetric
2. Trace Metals
3. Organic Solubles,
BaP, IR, NMR
4. Paraffins
5. SCHN (Particulate)
6. SCHNO (Organic Solubles)
7. Phenols and Nitrosamines
8. ERG Gravimetric
101
-------
TABLE 2 (Cont'd). TEST PLAN AND DATA MATRDC FOR EACH ENGINE
Fuel(s)
A and C
A + DII-2
and
C + 8005
Operating Condition
Speed
Idle
Peak Torque
Peak Torque
Peak Torque
Peak Torque
Peak Torque
Rated
Rated
Rated
Rated
Rated
Load, %
• • •*
0
25
50
75
100
0
25
50
75
100
Composite
Idle
Peak Torque
Peak Torque
Peak Torque
Peak Torque
Peak Torque
Rated
Rated
Rated
Rated
Rated
K _ M
0
25
50
75
100
0
25
50
75
100
Composite
Analysis Codes* by
Sampling System
4v 47
A. a 1
1258
1258
1 8
1258
1 8
1258
1258
1 8
1258
1 8
1258
1258
1258
1258
1
1258
1
1258
1258
1
1258
1
1258
1258
Hi-Vol
First Run
1346
1
1
1346
1
1
1
1
1
1
1
1346
1346
1
1
1
1
1
1
1
1
1
1346
1346
Rpt.
1 7
1 7
1 7
1 7
1 7
1 7
* Analysis Codes
1. Gravimetric
2. Trace Metals
3. Organic Solubles,
BaP, IR, NMR
4. Paraffins
5. SCHN (Particulate)
6. SCHNO (Organic Solubles)
7. Phenols and Nitrosamines
8. ERG Gravimetric
102
-------
3. 20 organic solubles, BaP, IRf NMR, paraffins and
SCHNO on the organic solubles, phenols, and nitrosamines
4. 56 SCHN on particulate
5. 60 ERG gravimetric.
A certain number of additional runs will have to be made, of course, to
allow for data errors, instrument failures, and so forth.
£. Gaseous Emissions, Smoke Opacity, and Bosch Smoke Numbers
for One Engine and One Fuel
To document the emissions of the 6L-71T engine, tests have been
conducted for both smoke and gaseous emissions. These tests were
intended to make certain that the engine is typical of its model and to
provide baseline smoke and gaseous emissions data on Fuel B (the fuel
specified in the Federal Register for diesel gaseous emissions and smoke
test purposes). The 13-mode steady-state smoke data by Bosch sampling
and PHS meter are shown in Table 3, and these data indicate low smoke
levels at all conditions. An additional set of full-load runs was made in
200 rpm increments from 2100 rpm to 1100 rpm; and the average PHS
smoke opacities were: 1.2 percent at 2100 rpm, 1.0 percent at 1900, 1.0
percent at 1700, 2.0 percent at 1500, 4.0 percent at 1300, and 11.8 per-
cent at 1100 rpm.
Gaseous emissions were also run on the 13-mode procedure, and
the cycle composite results are shown in Table 4. The computer tabula-
tions of the gaseous emissions results are given as Pages E-2 through
E-4 of Appendix E. These emissions agree well with certification data
on similar engines, so the major operational variables of the engine seem
to be in order. Transient smoke results were obtained using the Federal
Smoke Test Procedure and PHS smokemeter, and the transcribed data
sheets appear as Pages E-5 through E-7 of the Appendix. A summary of
these data is given in Table 5. Similar tests for gaseous emissions and
smoke will be performed using the other 5 fuels as permitted by the over-
all test schedule.
F. Dilution Tunnel Calibration
The blower used to pull dilution air and exhaust through the dilution
tunnel system was calibrated using a large laminar flow element and some
electric heaters, and the result is shown on page F-2 of Appendix F, with
supporting data and calculations on pages F-3 and F-4. It was necessary
to extrapolate the line back to a blower Ap of 3. 5 inches H2O because the
measurement system precluded readings with a Ap under 12 inches H2O.
The higher blower speed is being used to permit higher tunnel velocities
and consequent higher -.ample acquisition rates. The calibration with the
36:26 drive and 3. 5 inches HzO blower Ap yields tunnel bulk velocity down-
103
-------
TABLE 3. STEADY-STATE BASELINE SMOKE DATA FOR
6L-71T ENGINE ON FUEL "B" (EM-195-F)
PHS % Opacity
by Run Bosch Filter Reflectance Data by Run
Mode RPM
1 480
2 1600
3 1600
4 1600
5 1600
6 1600
7 480
8 2100
9 2100
10 2100
11 2100
12 2100
13 480
Load
25%
50%
75%
100%
100%
75%
50%
25%
— _ _
2
0.1
0.2
0.5
0.6
0.8
1.3
0.2
1.0
0.8
0.6
0.6
1.0
0.1
3
0.2
0.3
0.5
0.7
0.9
1.7
0.3
1.0
0.9
0.8
0.7
0.7
0.3
4
0.2
0.3
0.6
0.7
0.9
1.8
0.2
1.0
0.8
0.7
0.7
0.8
0.2
1A
0.2
0.1
0.2
0.4
0.6
1.0
0.2
0.4
0.2
0.3
0.2
0.4
0.2
IB
0.2
0.1
0.2
0.4
0.6
1.0
0.2
0.4
0.2
0.3
0.2
0.4
0.2
2A
0.1
0.2
0.2
0.4
0.6
0.5
0.2
0.4
0.3
0.4
0.3
0.4
0.2
2B
0.1
0.2
0.2
0.4
0.6
0.5
0.2
0.4
0.3
0.4
0.3
0.4
0.2
3A
0.1
0.1
0.2
0.4
0.4
1.0
0.2
0.4
0.4
0.4
0.2
0.4
0.1
3B
0.1
0.1
0.2
0.4
0.4
1.0
0.2
0.4
0.4
0.4
0.2
0.4
0.1
4A
0.2
0.2
0.2
0.4
0.8
1.4
0.2
0.4
0.4
0.4
0.2
0.5
0.2
4B
0.2
0.2
0.2
0.4
0.8
1.4
0.2
0.4
0.4
0.4
0.2
0.5
0.2
TABLE 4. SUMMARY OF 13-MODE CYCLE COMPOSITE GASEOUS
EMISSIONS FOR 6L-71T ENGINE ON FUEL "B" (EM-195-F)
Cycle Composite Emissions, g/hp-hr
Run
1-1
2-1
2-2
Avg.
HC
0.49
0.44
0.58
0.50
CO
2.66
2.33
2.10
2.36
NOV
11.6
11.7
12.6
12.0
HC + NOr
12.1
12.2
13.1
12.5
104
-------
TABLE 5. FEDERAL SMOKE TEST DATA FOR 6L-71T ENGINE, FUEL "B"
Percent Opacity by PHS Smokemeter
Run "a" Factor "b" Factor "c" Factor
1 11.9 1.4 20.6
2 12.8 2.1 20.8
3 13.0 1.9 21.8
Avg. 12.6 1.8 21.1
stream of the sampling point
v MI o\ blower revolutions (counts) s x
Vb = (11'9) - time, sec - TB PS
where the "S" station is the sample acquisition point and the "B" station
is the blower inlet. The constant in the equation for Vg^ (bulk velocity
at the sampling point) is 11. 9 for the 4 x 47 system and 12.4 for the hi-
vol system due to greater sample withdrawal rate).
Velocity profiles at the sampling station were acquired with a
Thermo Systems hot-film anemometer. The results were less precise
than anticipated, due to the influence of large-scale turbulence in the
duct. Averages over a number of runs gave usable values, however,
and the plots shown on Appendix page F-5 were constructed from these
data (included as Appendix F-6 through F-8). The positions as well as
the velocities in these profiles are referred to the duct centerline, and
although the "flat" sections of the profiles are not quite normal to the
axis, the deviation from centerline velocity (V^) is only about ± 2% in
the sampling zone. The reason for the higher velocities in the upper
right section of the tunnel is probably the overhead lighting which strikes
the tunnel from that direction. The tunnel can be insulated, or the lights
can be shaded if the profile distortion is felt to be a problem, but the
± 2% velocity gradient in the sampling zone is probably much less a
cause of anisokentic sampling than the large scale turbulence mentioned
earlier.
Temperature and concentration profiles were also taken (vertical
only), and the results are shown on pages F-9 and F-10 of Appendix F.
They are essentially flat so no comment is required.
G. Calibration of Sampling Systems and Engine Airflow Instrumentation
Measurement < Qow through the hi-vol sampling system (the one
which uses (nominal) 8 x 10 inch glass fiber filters) is performed by noting
the pressure drop tl.rjugh an orifice mounted about 0.79m (31 in) down-
stream of the sampling blower outlet in a 76mm (3 in) O. D. tube "tailpipe".
105
-------
Temperature is also measured at the orifice, so mass flow can be
calculated by the formula
mass flow = 5. 35
AP Pa
0.5
lbm/min= 2.43
0.5
kg/min
where pa is the atmospheric pressure in inches of Kg. The constant (5. 35)
was determined by calculation using the ASME flowmeter handbook pro-
cedure.
Flowrate measurements for the four-probe system using 47mm
filters are taken via flowmeters, and the calibration curves for the specific
meters employed are shown on page G-2 of Appendix G. Calculations and
data on which the flowmeter curves are based are given as pages G-4
through G-6 of Appendix G. These latter four pages also contain data and
calculations used to arrive at correction factors for readings of the dry
gas meters used to indicate total flow through each 47mm filter during the
sampling period. Flowmeter number 5 is used on the ERG sampler, and
it was calibrated in the same way as numbers 1-4. The calibration curve
for flowmeter 5 is included as page G-7.
The orifice used to measure engine air flow was calibrated against
a laminar flow element which has a calibration traceable to NBS standards.
The final equation (derived by applying the best squares method to the
logs of Ap and mass flow) is
Ma = 89.60 (Ap^)0-4842 Ibm/min = 40.64 JAp^)0'4842 kg/min,
and its derivation is given on pages G-8 through G-ll of Appendix G. Ex-
haust mass flow is simply air flow plus (directly measured) fuel flow.
H. Development of Simplified Operating Criteria for ERC Sampler
and Dilution Tunnel
After the ERC sampler was cleaned and made operable by a number
of minor corrections and repairs, it was decided that the instructions
supplied with it were not suitable for use by technical staff in the laboratory.
There is some question, as a matter of fact, that the operating procedure
as given in the instructions is workable at all. It was decided, therefore,
to review the sampler's principles of operation and devise a more usable
set of operating instructions. The instructions themselves appear as pages
H-2 through H-8 of Appendix H, and they are lucid enough for our technicians
to follow with only a little help. The calculations and considerations leading
to the instructions are presented as pages H-9 through H-16 of Appendix H,
and some simplified operating guides for the dilution tunnel itself are given
as pages H-17 and H-18.
106
-------
I.
Development of Mode Weighting Procedure for 13-Mode Tests
Obtaining a sample on one filter which is a true composite for
the 13-mode test requires that the total amount of raw exhaust gas
filtered in each mode be proportional to the product of engine exha\:st
mass flowrate and the time-based weighting factor for that mode. In
mathematical terms
mi =
E+D
i (Me)i or (Mor). = C2
E+D
where: i = individual mode, i = 1, 2, 3, , 13;
m^ =(m^) (time)i = total dilute exhaust filtered in mode i,
4 x 47 system, IbmJ
(Mor). = (M ). (time). = total dilute exhaust filtered in mode i,
hi-vol system, lbm;
(time)if = time in mode i, sec;
(Me)j = engine exhaust flowrate in mode i, lbm/min;
Wi = time-based weighting factor;
Ei = exhaust flow through tunnel, lbm/min; and
Dj = diluent flow through tunnel, Ib
Note that
so
(time)j
Ci =
E+D
Wi(Me)i =
E+D
Therefore, since both mj and (Mor)^ are essentially fixed by isokinetic
considerations, it is sufficient to use only one of the (time)i equations above
for computation purposes. The quantities IE+D i ^ (M_)., and (Sin-)
I f^ fi i i * w ± * -
are known or can be calculated from experimental data. If a value for any
(time)i is assumed, the constant C2 can be calculated and then the other
(time)i can also be calculated. To determine whether or not our choice
of C2 is reasonable, we can compute
(time) = y (timc)if
and choose a higher or lower value of C2 to make (time) more reasonable.
To minimize the complexity of the 13-mode test, it was decided
to determine the (E+D/E)^ with both the dump valves open (minimum ex-
haust backpressure). This decision means that we will not have the abso-
lute maximum particulate collection per unit time, but the latest experi-
mental data show that an adequate amount should be collected in a test of
107
-------
about 40 minutes' duration. It was also possible, of course, to deter-
mine (Me)j and a good average value for Mor while measuring the
dilution ratios. The data and some calculations are given in Table 6
TABLE 6. DATA AND CALCULATIONS
USED TO DETERMINE MODE WEIGHTS
i =
Mode
1
2
3
4
5
6
7
8
9
10
11
12
13
Wi
0.20/3
0.08
0.08
0.08
0.08
0.08
0.20/3
0.08
0.08
0.08
0.08
0.08
0.20/3
E+D\
E )i
32.2
22.6
19.8
20.4
19.4
18.1
31.5
12.9
14.4
15.8
16.3
17.6
27. 1
(Me)i
9.85
35.78
37.63
41.64
46.73
54. 14
9.85
74.43
67.45
60.36
54. 16
50.79
9.85
Time in Mode (min) by Assumption
1
2.00
6.12
5.64
6.43
6.86
7.42
1.96
7.27
7.35
7.22
6.68
6.76
1.68
TOTAL = (time) = | 73. 39
2
1.08
3.31
3.05
3.48
3.71
4.01
1.06
3.93
3.97
3.90
3.61
3.66
0.91
39.68
ASSUMPTION 1: (time)! = 2.00 min :. C2 = 0.1665
conclusion: (time) too long t\ assume smaller C2
ASSUMPTION 2:
C2 = 0.09
conclusion:
(time) OK, but combine 1, 7, and 13
to make one longer idle mode so tech-
nicians will have adequate time to
gather data
and on the lines just below Table 6, and the final schedule for the "11-
mode" runs is given in Table 7 (an "11-mode" is just a "13-mode" with
the 3 idle modes combined as mode 6). This schedule yields the desired
result, that is, weighting of modes so as to make a single filter repre-
sentative of a 13-mode test as that test is defined. If too small an amount
of particulate is collected during the test as scheduled, it can be repeated.
J.
Procedures for Data Reduction
Preparations discussed thus far have dealt with acquisition of good
samples by correct methods, analysis of samples to determine their com-
position, and the number of samples which should be taken to ensure that
an engine's entire range of particulate output is represented. This section
gives the final technical developments necessary to calculate engine total
108
-------
TABLE 7. WEIGHTING SCHEDULE FOR "11-MODE" COMPOSITE RUNS
Mode = i
1
2
3
4
5
6
7
8
9
10
11
Condition
Rpm
Peak Torque
ii n
M ii
II M
II II
Idle
Rated
n
n
n
n
Load, %
0
25
50
75
100
_
100
75
50
25
0
Mode Time,
Sec
198
183
208
222
240
182
235
238
234
216
219
Cumulative
Time, sec
198
381
589
811
1051
1233
1468
1706
1940
2156
2375
particulate output (and consequently the output of any species for which
analysis is conducted) from data obtained during the course of a test.
Data which are acquired during testing are perhaps best shown by the
data forms which are filled out by the people running the tests, so the
three types of forms are included as pages 1-2 through 1-4 of Appendix I.
For a given test, only half of each data form would be completed (either
top or bottom).
Mathematical development of data reduction procedures is given
as pages 1-5 and 1-6 of Appendix I, largely based on the results of cali-
brations discussed in section G and calculations discussed in section H.
Although the calculations are compact enough to be performed by hand
for a few cases, the large number of samples being taken for this pro-
ject make computer processing more economical in the long term. Ex-
amples of the encoding sheets from which data will be keypunched (12
data cards per test) are given as pages 1-7 and 1-8. The computer pro-
gram being used for the processing is included as pages 1-9 through 1-12,
and sample results are given on pages 1-13 and 1-14.
Possible Problems and Corrective Action
The only problem in view at this time is the continuing background
interference with the Phenol analysis, and it is hoped that the new "clean"
filters currently on order will eliminate the difficulty.
Plans for the Next Reporting Period
It is planned that all samples on fuels A, B, and C for the Detroit
Diesel engine, except possibly those for Phenol analysis, will have been
acquired. It is also planned that chemical analysis of all samples taken
109
-------
while using fuels B and C will be complete or well underway. The gravi
metric calculations (using the computer) should be well underway by the
end of the next reporting period.
Submitted by:
Charles T. Hare Karl s Springer
Manager, Advanced Technology Director
Department of Emissions Research Department of Emissions Research
110
-------
APPENDIX A
EQUIPMENT DESIGN SKETCHES
111
-------
619mm
-(24.
2.44m
... . (B lt J ^
838mm
" (35 in) "|
i 1
^^ ??Qmm fQ i n^ /I Crv~,~»
(0 LI)
6 9 9 mm
i
I
T"^ "1 4- ^^^^^^^*^fc_ '
Diam.
AiF
(17.7 in) I. D.
Exhaust
_, /0 . > — ^— Mixing Orifice
76mm (3 in) O. D. e
Raw Exhaust
Transfer Tube
Dilution Air Filter Enclosure
Hi-Vol Sample Probe
127mr
(Sin) O.
LL
in)H
SCHEMATIC SECTION OF DILUTION TUNNEL FOR DIESEL PARTICULATE SAMPLING
-------
MAKE 4- FAC* FIZOOV
ST A IK/LESS ~T«EU
IV-2>716-001 HAKE
STRAIGHT
BE O.SO^
TO 8.E READ TO 2>
CT»P)
fitnrtfAjslOMS NOT CPlTlCAL '.
-\AffR. ».S t» z.Q t
eooy >.c n z.o
0,S t.
TA,ptR£
-------
C. i/V)
t40T TO
*-
•roLtehwc.fi
10.01
NOTED
1. S.TAKOLESS, S.TEEL
MAHE. 1 EACH
O.SO 01A
0.011
1.0
IS* TAPtR
SEC.TIOM c-c.
na
0.41) *»;
o.ooi
O.OI •<» 6.O2.
-------
APPENDIX B
FUEL SPECIFICATIONS AND INSPECTION
115
-------
DIESEL FUEL SPECIFICATIONS
SWRI PROJECT 11-3718-001
Emissions Test Type 2-D
Test Fuel A
FLA Analysis
Aromatics, %
Olefins, %
Saturates, %
Distillation
IBP, °F
10%. °F
50%, °F
90%, °F
EP, °F
Sulfur, %
Nitrogen, %
Cetane
Gravity, °API
Flash Point, °F
Viscosity, cs
Max
Min
27(D
Remainder
Max
10
2
95
Min
5
-
88
Blend 1
8
1
91
400
460
540
610
660
0.5
50
37
3.2
340
400
470
550
580 , m
0. 2<2>
42
33
130
2.0
310
340
400
480
500
0.05
0.02
60
55
240
320
380
440
480
48
50
120
1.5
320
331
379
456
492
.007
46
47.4
Test
Desired
20
1
79
about 370
430-440
480±20
550 max
580
0. 05 max
40 min
40
Fuel C
Blend
10-11-73
20
1
79
432
449
463
508
580
47.5
38.7
2 min
(1) Must not exceed 35 percent for Project 11-3718-001
(2) Should be around 0. 3 percent for Project 11-3718-001
-------
ANALYSIS OF FUELS TO BE USED DURING DIESEL
PARTICULATE MEASUREMENT PROJECT
Project Fuel Cose A B C
SwRI Fuel Code EM-197-F EM-195-F EM-198-F
Fuel Type No. 1 Kerosene 2D Emissions "No. 1-1/2"
Distillation, °F
IBP 330 384 410
10% 354 434 426
20% 360 460 441
30% 366 483 446
40% 376 500 450
50% 387 518 455
60% 402 531 460
70% 418 548 465
80% 437 569 479
90% 460 601 502
EP% 525 673 594
% Recovery 99 98 99
% Residue 1 2 1
% Loss 0 0 o
Aromatics, % 9.2 35.1 23.0
Olefins, % 0.8 0.0 1.0
Saturates, % 90.0 64.9 76.0
Gravity, °API 46.8 34.9 39.4
Cetane (Calculated) 51.0 49.5 49.5
Total Sulfur, % 0.003 0.319 0.010
Weight % C 85.0 86.5 85.3
Weight % H 13.9 12.8 13.9
Weight % N 0.08 0.10 0.04
Viscosity, cs 1.62 2.7 2.12
FlashPoint, °F 130 176 185 +
117
-------
APPENDIX C
SAMPLE NMR AND IR SPECTRA RUN
ON EARLY ORGANIC SOLUBLES SAMPLES
118
-------
11
11
4GCO
3iOO
3000
WAVENUMBER (CM'1)
2500
2000
1800
1600
1400 1200
WAVENUMBER (CM ')
1000
800
625
SAMPLE
ORIGIN.
SOt VENT ___0fllti
CONCENTRATION JWE_
CEU PATH nHH_
REFERENCE ___^_--
O -O- -O- --Q- 0
REMARKS
SCAN SPEED ,
SLIT M.
PCRKIN ELMER
PART NO. 477-SPOT
OPERATOR,
DATE
REF. No. _
n
a
-G
-------
-------
APPENDIX D
PROCEDURES AND EARLY SAMPLE DATA FOR
PHENOL AND NITROSAMINE ANALYSIS
121
-------
Method for Determination of Phenols and N-Dimethylnitrosamine
in Particulate Matter Collected on Glass Fiber Filter
1. Cut filter in pieces approximately 5 x 40 mm and place in
200-ml round bottom distillation tlask.
2. Add 70 ml of 1% H3PC>4 in water.
3. Connect distillation flask to vertically mounted small diameter
(8 mm O. D.) water cooled condensing tube. This distillation set-
up is similar to a Kjeldahl distillation apparatus.
4. Place 5 ml of 50% KOH in 50 ml beaker and place beaker so that
outlet end of condenser tube is immersed in KOH solution.
5. Distill over 35 ml HzO and rinse condenser tube with 5 ml HzO.
Should now be approximately 45 ml in beaker.
6. Transfer, without rinsing, contents of beaker to 125-ml separa-
tory funnel.
7. Add 13 gm NaCl to funnel and shake to dissolve.
8. Rinse condenser Lube with 10 ml benzene and collect in 50 ml beaker.
9. '"'ransfer benzene to separatory funnel containing distillate and shake
vigorously for 1 minute.
10. Drain aqueous phase into another 125-ml separatory funnel. Discard
benzene.
11. Add 10 ml dichloromethane (DCM) to separatory funnel containing
aqueous phase and shake vigorously for 1 minute.
12. Collect DCM in small vial and save.
122
-------
13. Add 10-ml hexane to separatory funnel and shake well.
14. Drain aqueous phase into 100-ml volumetric flask. Discard
hexane.
15. Add 1 drop Phenolphthale in Indicator Solution to aqueous phase.
16. Add concentrated H3PO4 to aqueous phase to indicator end-
point then add 2-3 drops excess t^PO^
17. Cool to room temperature and add 0. 5 ml diisopropyl ether (DIE).
18. Shake vigorously for 1 minute and immediately pour into 50-ml
volumetric flask using appropriate funnel.
19. Swirl contents of stoppered flask and then allow DIE to collect on
aqueous surface in neck of flask.
20. Insert ground glass stopper, to which has been attached a short
length (60 mm) of 2-mm I. D. capillary tubing, into mating glass
joint on flask.
21. Using a syringe and needle, inject water into flask through pre-
viously inserted silicons plug in flask body, so as to force the
DIE up into the capillary tube.
22. Using a micro syringe, withdraw 5/fX of DIE and inject into gas
chromatograph for analysis of phenols.
23. The DCM previously saved is transfered to a micro concentrator
and evaporated down to 0. 5-0. 75 ml.
24. 20xf£ of the concentrate DCM extract is injected into a gas
chromatograph equipped with an Electrolytic Conductivity Detector
used in the pyrolitic mode for selective detection of N-nitrosamines.
123
-------
CHROMATOGRAPHIC CONDITIONS
Column:
Column Temp:
Detector:
Detector Lens:
Phenols
6' 10% OV-3 + 1% FFAP on 80-100 mesh
Gas-Chrom Q-AWDMS
125°C
FID
16X
Dimethyinitrosamine
Column:
Column Temp:
Detector:
Detector Lens:
6' 10% Carbowax 1540 + 10% KOH
on 60-80 Gas-Chrom Q
125°C
Electrolytic Conductivity (Pyrolytic mode)
IX
124
-------
/ us o-f-
-------
i
!:
•9-
'/
Zx.Tro.tT
I i
. ! . ' i •_•
\).*t/se J ...F.i'/Ter_ 9 x /o"
i
. i'~~
A- --* 1-
-------
1 .
Ill'
, . ,
Li-:;.
I '
. ....'.„
, •
..
1 .
i .
i
....'!...;.
i
1
-z-<
,,*«
•
-• m
: •
i
i KIW U-
-
f-
-------
-------
-------
1
U)
0
1
•
i
1
1
-ft
i
Jj
r
\
2
B
— ^— 1
. . .
•
i
i
•
r
V
TXTH
D — ~~
•
....
• !
r-j-i-
!.!
-*~~*
-
' ' 1
i . !
O
' i •
' ' ! I
.: • •
f 1
^ '
. • i
rj .
rXji .
^jfi
/
i ' '
V
V
. I • )• •
/ •
1 ' '
: 1 ; .
: : ; .
• * ; 1
t
: ' ' ]
i i .•
: . . - 1-
O
vj
i
i r .
^ -pi- •
"J
"*^
^B]
i i '
i
j .
1 f
i ! !
! t
• ; '
, 1 !
i I i
' 1 ! ]
'
•'ii
' ' 1
-
'-
,
!
!
•j
!
' !
•i
.
-
:
-
1 i t i
+- -i f--
1
1— - - -
1 .
-t-H
! '
r
r n
1
;
1 1
-H
; t
TT^
1 :
— : r---
t ; ! '-
r . i i
i " -
IT
O' T
!
•Ite
H-r-i
IT" '
Tin
fl-I-t-
i i
1 i
T T~f 1
rT"
•-j~-
m
| !:
•
'
' '
^J
1
•H
-
--
^
'
-
-
;
-
-
_
-
-
-
:
r
J
:
-
;
-•
.
1
:
•
•
•
".']
i
-
"
t-
:
-
-
;
;
-
i—
•
...,
-
-
n
-
..
•-
I
-
•
—
:
-
--
r-
I
..
—
r
•^
~>
-
i
•
'
•-
-
"
'
::
•
--
-
—
-
-•
•1
-
-
-
:
-•
~
•
-
~
-
-
'
-
-
-
.
-
-i
-
-
_
-
•-
—
— i
;i
-
•
-
-
-
i—
j
—
.
-
-
:
_
__
:
E
-
T
I
-
:
-1
-
-
- -i
-
-
-
•
I
I
-
-
-
:
-
•
-.
'-
•--
-
-
•
-
•
i
-
,.
-
-
-
—
:
-
•
--
-
1
:
— i—
!
-j
j
1
|
'
:ii
"-pi
±
1
i
i — i —
-.^ -.
j
r i
-
i
~*
—
-•*•-
H
b.b.
•
I
;
-
-T-rH-M— -* i- ..»-..-•:
! ' i ! ' : ', ' ' '
• • l -]••,•••'
i , -T I 1 i i -t- i i— »-« I- • •' -•
i ' «« i
i ' i
-O
i ' 1
fc V "^-s
> .IT
' i I» <*>
O i iS
,yfe ^ ^ .
^«O ' 1 ** • '"Ti 1
^ ^ Xl
V* f '•t rt C1
v" •* w ^» ^
* c? ^
^ V» ^
Vi ? •*• " '
C> > . ; S i J
iv r%
! ^ ! 5
*• ^ , <* 1 •••
i
i ' ' '' '• ' ! ' 1
|-j- -j-fi t •• H-rr 'T '
•» rrr -IT 'T T] • ^ ----- •
: ,..j 1 -4--U-- TTII .; .... i
JffiffifiiSMs
-------
-------
APPENDIX E
GASEOUS EMISSIONS AND SMOKE DATA
ON THE DETROIT DIESEL 6L-71T ENGINE
USING FUEL B
132
-------
13-MOOE FEDERAL DIESEL EMISSION CYCLE
PROJECT:ll-3718-001 DATE OF TEST 1-3-7* TEST NO.l RUN NO.J
DETROIT DIESEL bL-7lT FUEL TYPE EM-1S5-F
MODE
1
2
3
t
5
b
7
8
q
10
11
12
13
ENGINE
SPEED
RPM
480
IbOU
IbUU
IbOO
IbUO
IbQO
ISO
2100
21UO
21UO
21UO
21UO
f80
TORQUE
LB-FT
0.0
IS. 8
ISO. 8
381. fa
57H.2
7b5.0
0.0
bflq.B
518.2
3ff .q
173.3
If .0
0.0
POwEft
BMP
0.0
f.8
58.1
Ub.3
17«t.q
2:13.1
0.0
275.8
207.2
137. q
bS.3
S.b
0.0
FUEL
FLOW
LB/MIN
.03
.18
.Hb
.78
1.1?
1.50
.Of
1.B2
1.3q
1.01
.bf
.33
.05
AlK
FLOW
LB/MIN
q.Hf
3H.20
3b.qS
3R.80
tf.35
HS.ID
H.f*
b7.qq
bl.70
55. 8b
51.10
f7.q3
q.tS
EXHAUST
FLOW
LB/MIN
q.f 8
31.38
37. ff
tO. 5 8
H5.H7
50. bl
q.ts
b^.Bl
b3.10
5b.B7
51. 7f
f8.2b
S.50
FULL
AIR
RATIO
.00*
.005
.013
.020
.025
.031
.005
.027
.023
.018
.013
.007
.005
HODE HC
PPM
^
e
3
•t
S
b
7
8
q
10
11
if?
13
CYCLE
113
121
lOb
bf
7b
82
bO
108
9b
7f
80
100
92
CO*
PPM
122
130
70
S3
175
138b
IIS
2fa3
100
82
78
15
12b
COMPOSITE
BSHC -f
NO-H- WEIGHTED
PPM BHP
178
8fa
282
580
S71
11SH
201
12b2
R28
552
28H
112
iq«f
BSHC =
BSCO+ =
BSN02++=
BSN02++=
0.00
.38
f .t>5
9.30
13. ss
18.fa5
o.oo
22. Ob
lb.58
11.03
5.5
-------
13-MODE FEDERAL DIESEL EMISSION CYCLE
PROJECT:ll-3?lB-001 DATE OF TEST 1-5-7* TEST NO.2 RUN NO.l
DETROIT DIESEL bL-7lT FUEL TYPE EM-195-F
MODE
1
2
3
4
5
b
7
8
9
ID
11
12
13
ENGINE
SPEED
RPM
180
IbOO
IbUO
IbUO
IbOC
IbUO
4Su
21DO
21UO
2100
2100
2100
480
TORQUE
LB-FT
0.0
1H.O
192. b
383. f
57b.O
7b8.5
0.0
702.0
527.0
351. S
175.1
14.0
0.0
POwER
BMP
0.0
4.3
58.7
lib. 8
175.5
234 .1
0.0
280.7
210.7
140.7
70.0
5.b
0.0
FUEL
FLOw
LB/MIN
.05
.18
.47
.79
1.11
1.51
.Of
1.82
1.39
1.02
.bS
.34
.03
AIR
FLOW
LB/MIN
9.4b
33.79
37.42
40.72
45.17
52.31
9.44
b9.93
b2.8b
57.51
52.«fO
47. 9Y
9.11
EXHAUST
FLOW
LB/MIN
1.51
33.97
37.89
41.51
4b.28
53.82
9.48
71.75
b4.25
58.53
53.05
48.28
9. If
FUEL
AIR
RATIO
.005
.005
.013
.019
.024
.029
.DO1*
.02b
.02?
.018
,ni?
.007
.004
MODE riC
PPM
1
2
3
•*
5
b
7
8
9
10
11
12
13
CYCLE
128
148
lib
5b
40
32
72
72
bO
88
104
120
124
CO*
PPM
153
157
84
77
14?
1054
112
2b4
92
70
b4
8b
120
COMPOSITE
BSHC +
NO++ WEIGHTED
PPM BHP
155
9b
285
5bO
903
1119
IbO
1300
954
5b?
272
135
13b
BSHC =
BSCO+ =
BSN02++=
BSN02++=
0
4
9
14
18
0
22
Ib
11
5
0
2
11
12
.00
.34
.b9
.34
.04
.73
.00
,4b
.8b
.2b
.bO
.45
.00
.441
.328
.709
.150
BSHC BSCO+
G/HP HR G/HP HR
R
15.55
*99
.Sb
.14
.10
ft
.24
.24
.48
1.04
13. bb
R
GRAM/BHP
GRAM/BHP
GRAM/BHP
GRAM/BHP
32.
1.
•
1.
b.
1.
.
.
1.
IS.
HR
HR
HR
HR
R
93
43
72
02
37
R
77
74
77
27
53
R
BSN02++
G/HP HR
32.
7.
8.
10.
11.
14.
12.
10.
8.
50.
R
91
94
bO
29
11
R
3b
57
18
90
34
R
HUM.
GR/LB
27.1
27.1
27.1
27.1
27.1
27.1
28.8
28.8
28.8
28.8
28.8
28.8
28.8
CONVERTED TO WET BASIS
CONVERTED TO WET BASIS AND CORRECTED TO 75 GRAINS
WATER PER LB. DRY AIR
134
-------
13-hODE FEDERAL DIESEL EMISSION CYCLE.
PROJFCT:ll-3718-001 DATE OF TEST 1-5-7* TEST NO.2 RUN NO.?
DETROIT DIESEL bL-711 FUEL TYPE EM-195-F
MODE
1
2
3
4
5
b
7
8
c,
10
11
12
13
FNGINE
SPEED
RPM
HBO
IbUO
IbUO
IbUG
IfallO
IbUO
480
21UO
2100
2100
21l'0
21L-0
450
TORQUE
LB-FT
0.0
14.0
192. b
383. 4
57b.O
7bb.8
0.0
703.8
528.7
351.9
17b.8
14.0
0.0
POwER
BHP
0.0
4.3
58.7
lib. 8
175.5
233. fa
0.0
281.4
211.4
140.7
70.7
S.b
0.0
FUEL
FLOW
LB/MIN
.03
.18
.4?
.79
1.12
1.50
.03
1.81
1.40
1.03
.b?
.33
.03
AIR
FLOW
LB/MIN
1.07
34.39
37.37
40.52
45.03
51. 49
9. Ob
70.35
b2.bb
Sb.83
52.23
47.84
9. OS
EXHAUST
FLOW
LB/MIN
9.10
34.58
37.84
41.31
4b,14
52.98
9.09
72. lb
b4.0b
57. 8fa
5?. 90
48.17
9.12
FUFL
AIR
RATIO
.004
.005
.012
.020
.025
.029
.003
.02b
.022
.018
.013
.007
.004
MODE HC
PPM
1
d
3
u
5
h
7
a
s
in
li
is
1 ^
CYCLE
112
IbO
118
84
88
9b
128
104
92
103
112
13b
13b
CO*
PPM
75
120
94
52
135
1052
91
24b
b8
44
39
b4
99
COMPOSITE
BSHC +
NO++ WEIGHTED
PPM BHP
152
84
294
bOb
1010
1238
145
1410
1U19
b04
301
104
152
BSHC =
BSCO+ =
BSN02++=
BSN02++=
0
4
q
14
18
0
22
lb
11
5
0
2
12
13
.00
.34
.b9
.34
.04
,b9
.00
.51
.91
.2b
.bb
.45
.00
.583
.098
.5bl
.144
BSHC BSCO+
G/HP HR G/HP HR
ft
17.12
1.00
.39
.31
.29
R
.35
.37
.5b
1.11
15.44
K
GRAM/BHP
GRAM/BHP
GRAM/BHP
GRAM/BHP
25.
1.
m
.
b.
1.
.
.
,
14.
HR
HR
HR
HR
R
50
59
49
93
27
R
bb
54
48
7fa
37
R
BSN02++
G/HP HR
29.
8.
9.
11.
12.
IS.
13.
10.
9.
38.
R
50
20
2b
48
13
R
bl
34
74
73
b5
R
HUH.
GR/LB
37.4
37.4
37.4
37.4
37.5
37.5
37.5
39.1
39.1
37.5
37.5
39.1
39.1
CONVERTED TO WET BASIS
CONVERTED TO WET BA3io AND CORRECTED TO 75 GRAINS
WATER PER LB. DRY AIR
135
-------
FEDERAL SMOKE TRACE EVALUATION
Vehicle
Date /- /.
Evaluated by
Engine Model <£/.- 7/7"
Run No. /
Accelerations
First Sequence Second Sequence
Third Sequence
Interval No. Smoke % Interval No. Smoke % Interval No. Smoke %
/
.2
3
4
£*
^
7
9
/o
/2*
/3
7+
/g~
Total Smoke
Factor (a) = ,
f. ;Zx
=2/. ^*"
^^ ^
*?3 ^?
-L0.f
«?/. ^
/£.3
M.o
J+.8
/3,0
/I.D
-? 5^
/ /
/ s*
/tf
% vZ^/.tf
&4.t
1
^2
J
^
£*
(,
1
/
/a
t3s
/3
f/L
/£*
,/.f*
jr ^ —
A? ^
/fS^
jo.O
/$.&
/6.S~
/L.-S"'
/3.D
H.0
//.a
0s-
^.3^
^.5—
-
13
/s —
^f^
yV. 4
/f.^7
/9 f
JS.S*
^&. •*5'
*^» »^
/V7^"
/3 0
//.O
9-o
6.3
*3-0
L >^
/.r
/tit.
Third Sequence
Interval No. Smoke % Interval No. Smoke % Interval No. Smoke %
/
^
J
4
,5*"
Total Smoke
Factor (b) =
/ a
/ ^
/jj,
/^
/•3~
/
.a.
^/
y
^
/ ^
/>£"
/ 5^
/.S^
/.sr
% »^ ^ 5^ •S*^
^V » X^«
15
Peak Readings
First Sequence
Second Sequence
/
^,
*3
y
,r~
/ g**
/ ^"
/ ^^
/ y
/ ^— "
i.f~
Third Sequence
Interval No. Smoke % Interval No. Smoke % Interval No. Smoke %
/
-2-
^
-££.£>
*23.D
^2./.S*~
/
^
^f
<=20.0 II /
/f- ^* '2>
/J-^ I ^
/?. S
rt.o
/S S~~
Total Smoke % 70,S~~ +5~7.& sf- &
Factor (c)
136
-------
FEDERAL SMOKE TRACE EVALUATION
Vehicle
Engine Model
Accelerations
First Sequence
Date
/. - 7 / 7~
Evaluated by
Run No.
Second Sequence
Third Sequence
Interval No. Smoke % Interval No. Smoke % Interval No. Smoke "I
1
«£
J
y
g*
£
7
^
/o
/2*
/3
/*J.
IS"
4. 6
/-?.5^
j./. j —
^6.0
J.0.O
-za.s^
31. O
i&.O
ll.O
t^.O
f.f'
3,5"
^2-S^
J.f
1
a.
3
*J.
f
c>
1
t
q
/o
II
fJU
/3
,+
s'.o
J^.O
£/. S~
£3.0
/.Q
/c? ^
//.a
£. D
4s~
4 ' £>
3 sr-
Total Smoke %
Factor (a) =
Lugging
First Sequence
Interval No. Smoke % Interval No. Smoke % Interval No. Smoke %
Second Sequence
Third Sequence
Total Smoke % //.
Factor (b) *
//.f
15
Peak Readings
First Sequence
Interval No. Smoke % Interval No. Smoke
Second Sequence
Third Sequence
Interval No. Smoke %
J__
. O
.2,
. 0
Total Smoke % £,3. O
Factor (c) -
137
-------
FEDERAL SMOKE TRACE EVALUATION
Vehicle
Date
/- /.
Evaluated by
Engine Model <£/.- TV ~7~~
Run No. ^
Accelerations
First Sequence
Second Sequence
Third Sequence
Interval No. Smoke % Interval No. Smoke % Interval No. Smoke %
,
^
j
4.
^
6
7
ft
£2
SO
//
SI,
/*/,
/s~
to
J>Jf£)
-Z3 S~~
/f. 0
/9.o
//.^
/&> s~
/L.O
/4.0
//.x
/.5"~
&.£~
^ £,
~2.6,
J.6>
y
.2,
,3
y
^*
^
7
/
1
/£>
//
/£>
/J
/y
/f
S.O
/&>. O
£~"
/S'.f)
/3.A
/o.o
3*P
-2-7
^3.3*
3.0
/
3,
3
4
to
7
/
f
JO
/ /
/.£/
/3
/*?.
/£*~
/S'.O
cZS'.O
£^.0
<23..O
j?t>.0
tf-3
/7-0
/7-O
/V.5^
/2.O
1.0
b.f
j2..f^
^•<^
«=?./
Total Smoke % /?/. /
Factor (a) =
45
Lugging
First Sequence
Interval No. Smoke % Interval No. Smoke % Interval No. Smoke %
Second Sequence
Third Sequence
. A
Total Smoke
Factor (b) =
15
Peak Readings
First Sequence Second Sequence Third Sequence
Interval No. Smoke % Interval No. Smoke % Interval No. Smoke %
/
^
*3
^3.S~~
-Z/.0
/f.O
/
*Z
\3
*9V.0
z$. &
/7.5 —
/
^
3
-2^.0
*ISJ.0
^*.O
Total Smoke % &J.S~ 6>0.$ 7«2.<9
Factor (c) -
138
-------
APPENDIX F
CALIBRATION AND PROFILE ANALYSIS
OF DILUTION TUNNEL
139
-------
CAU &RAT iow
140
-------
PULLEY*. BLOUl&IL PUULEV
TAt&ET
-IS
IS"
7S_
too
IPO
JO 0_
J1F_
125
.ȣ?
IS*
.-75".
J00_
16«»
AC.TUAL.
J?5_
74
. 2*_.
.1?9.
il I
itt
BMXUt-R-
IT-
70
T)
IZ
no
\ to
BLOWER.
Q.OMCOR.R..
520.
50>.
H.34-
11.2.7
Si?.
tUZX)
ST-3.
555.
540.
S14.
J0tl_
Iti!
5 VI.
5 10.
0-97^1
0.9T1-8
O.J753
0,914-e
0-91SI
CORR.
o.blfct
O.&UC,
JL'JJL^J jJLI^lrL
A«»T5>_i^L!!'.Ll_
0.9758 I 0.7569
I
Id tO I 517.
0-9748
_0_.9_7_SJ_!
0.97SI '
Q. CORt.
498,
413,
449.
4*1.
6'-Ouj
y/x> »
/ft.
_p_. ^ls
0.312
J:^7-_
.*2-7-
413.
~r~
i
494,
467.
400-
441.
O.ilS
.Qi.
_o-_.
o.
p.
p. a\9
0. 3 IS
O.iZQ
0.317
»7 r
5C.4,
0-9746 j O.bl7?
450.
10
sn.
Pfc&^.SOg.g. ^9. 44-
__
O
141
-------
5(c
TUMQEU BLQtU Eg. .C.AUB RATIO A3
ftLOtfElt -WLE1
TAtGCT
75
IS
75
100
l 00
1 00
tlS
l2S
,TS-
ISO
ISO
ISO
75
75-
100
too
l*S
IZS
itr*P.,«F
AtT ITAL-
IC
78
7t
97
9C
,c
• is
nft
il9
149
142
139
15
98
<)C
M9
lift
&MHUE.R.
Afr.^^0
«0
15
ZO
\0
IS
ZJ>
10
IS
ZO
10
ir
10
10
IS
IS
ZO
10
z»
BLOWER. LArWi Uf\
Rty/-,^ Q.owco^e..
1334- 4ZC.
1330 414.
I32.S 39B
j
I33C. j 44 \.
»332 ; 42-8.
!
13 1C ' 415.
1337 458.
1332 447.
I31B 431.
• 1.3C. 48\.
i-bil 457.
1327 440.
133C 4\9.
1332 ' 4«>7.
133 1 4^7.
I3T5 ! 4-1 0.
>33S 453-
1317 47,2..
H FLOtU
t> CoR-fc.
0.91(«4
o.^t?
0.9171
9.
370.
3 57-
3S9.
3S3.
344.
404.
39Z,
361-
%fc7.
390.
3S(p.
"2.9.4Z. J^ H,
142
-------
-------
7/IT/
4.
&.(.
B.I
l-l
t.l
£.1
4.1
3.1
2,t
I.I
0,1
-0.4
-1.4
-1.4
-3.4
-4,4
•4
-C.4
-1.4
-7.%
V, ^
Ru* t
s-4
s.ss
t.t
J..3
t.4
(..5
U.S
U.5S-
fc.l
t.L
<.,tS
I. OS
t.o
1.1
S.&
fc.o
s-u
S.ss
S-*5
/MC
R«*i
9.S
(0,0
10.5
\\.o
10.%
10,1
—
.0.2,
—
10.4
—
10. 1
9.1
9.5
9.1
8.&
V
R^l
O.fttl
O.t>9l
o.»79
l.OVl
i.on
1.04-3
1.043
1.01)
0-m
0.9^5
1.001.
0.971
0-U3
0.919
0.931
,
0,j\aj
0-899
o.«9l
0.8SJ
'A/t
Ru.l
0-915
o,m
I.O.I
1.0 00
1.040
I.OM
—
o-99L
—
1.001
—
0,965
—
0.954-
o.b9(,
O.fl9(,
0.84B
*«y
O.o9l
o.9t7
0-99U
I.03C,
1-034-
l,04i
1.02,7
1.019
0-9H
0-995
1.003
o-9ll
0-913
0-919
0.9&3
-
0.™\i3
O.b9l
0.%93
O.B&3
«eU«i
8,1
8.1
l-l
l.l
E.t
4.t
VI
i.U
M
0.1
-0.4
-4.4
-2A
-5.4
-4.4
-S.4
-1.4
-1.4
-l.B
V,
*.n2.
5.45-
1.05
1.4-
8.9
8.9
9.0
95
9-4
9.1
9.4
16.0
10.7.
IO.J
9.8
10.1
(0.4
io.g
10,0
^*f ItJ
ftu^4
(o.l
5.5-
9-0
9.L
).l
—
10.0
__
9-9
—
tO.O
io,!>
____
10.1,
to, 4
9.8
10,5
c~
EMKS
4.0
S.4
5.5
s.u
&.4
S.I
5,5
—
S.5
• —
(..0
s.9
"
t.\
S,B
g.^
E«l
6.SS&
0.7Z1
0.758
0.911
o.v*
0-911
0.97%
o-9l^
0-914-
0.9t3
1,01?
I.04S
I.05S
1.004-
1-OlVi
.045
I. Oil
1.011
I.OZ?
v/
e«* 4
o.oti
o.^sz
o,nz
0.9C1
0-912
1. 001
—
0-391
—
1,001
—
I • 0^}^
—
1 fO^Z"
""•
I-04Z-
0.9&1
1.001.
s/t
K.«S
o.u«7
o>na
o-ns
o.9t>l
0-9Z8
—
O.S9S
—
0.9*5
—
1.014
—
(.031
—
1.014-
I.04&
0.^91
0-9U
Av^.
0.&I9
o.n4
o,«(,8
0.94^
0.914-
O.ni
0-9SC,
o.H^
0.974-
O.U3
1.014-
\.04S-
1.03,9
1-004
',041
1 nACT
I.Oq-S
1
l.0£l
).0>6
0.919
.
r
10.
9.11 9.9ft
"T -x
I 14
, uAJL 4jo^t >a «^flA*d_*=?_ -»A£i _
dU^ btjiO
-------
- OCM
U -
Q -
PR»FUE
U
(•.(.
s.fc
4.(.
3.1
r.t
«-L
o.l
-0X1
-..*
-2.4
-J.4
-4.4
-C.4
-t.4
t.ost
1.0*4
1.043
1.0*7
1,01*
0.994
0-9*
1.103
0.911
0.915
0.979
0.913
0-913
o.«)7
(..I
S.C
4-.L
1.\.
z.t
«.t
O.I
-«.*
-1.4
-Z.4
-V4
-4.4
-S.4
-C.4-
0-WS
O.H«V
0.9tl
0.9SI
Oi,fc5
0.974
0.9 1 3
l-Old
1.045-
I.OJ)
I.Oo
-------
f
of
_o
TOP
8.86
_?IU_
ft. I
_0_
0.&9I
6.85
7.U2t
7.50ft?
?.
-O.S-
0.651
-0,4
*^»V%-
ft .6 73
J7A.
G.I
4.1
J.OSt
JI.OS4/
11*7_
l.0\)
7.SC.9C
\-^^»l
(,.B}7l
4rO-
Tr
-i.-O—k-r5«4»-
-frO-
5.
^A^JP-
5.1061
4rO —
lr«H
Ir^ltJ
-i.
trO-
»7<^
4~U,S£-
M19
^_,,,^
2.4
0-71J
. 15 SI
I.I
_o-t_
0
!•*
o.-V
0^)5
1.000
!^?7_
o
-OiW—
-OH-19H-
9 i??L
1.003
1.000
O^Ji
0
4-rftHJ—j—4,0—
4,« -
O.ftbOi -4.0
-o.wot-- -0.4
1.
o-oftot
= ln.Bjje
•«.) - 0.8U^
Vr
6.85
B.U
O.UI)
5.U14-
-OrS
M^_
-7-8
4W-
7.t
-Vr«'
4^*-^
_?!*_
t*
-I*S—
0.4-
J.oi8_
l.OSZ
?.«&»?
-^.
t-lSTO
5.1)04.
0.122
S.l&V?
-4-.3&KB-
*««4i4
IwO-
5.4
1-045
S.C.430
-4^*-
-4w*-
4.4-
4.SB04-
-».«--
H.^-
-4^y—
1.0*4
0.9St
1.030
-------
:
-
"i~
-^-6-
t-r
$
m
~ .. I
:; I
'—: I
#4-2r
.'. ~ ' 1^.7™
1
4-t
-• 1.
rr'|'.
EEfEE:
=3
Ic
-jrlr}: — :
T^r
i
-? _t _4
• »> BEUMO
J
: ' •[ ' j
___] LI
.-H_4-4
-|--,
147
-------
.... t
lit
z^z-|~::
... .
r—J-——
T.T-."J '.". ~
is=
tM
_¥L_
t&UT OP
ibc M»
tOW^POTtAJTlOM
^WT-
-A&
4
-H-
T
148
-------
APPENDIX G
CALIBRATION OF SAMPLING SYSTEMS
AND ENGINE AIRFLOW ORIFICE
149
-------
150
-------
DAT t 1
Of FLOtU
FU.T.)
1
-h
H*
1,
? wtit*-
- C.
DK-V
PAPT OP sviT6«vi TO
tAuB."
rvo". mifi'C. PR I--ORE = to.- Z.J. 14" •*..
METE.IL Mo. _L
• t |1.
( lit.I ( z.s j i2_ o._i4S! 64 | 5
! '° ?l2- _. l?|l?_:_0 i_?_i
1 30 .1 o.B
— «J
£T) I • 4*
10_ 2.2
90 i-J
0-2-
12.B.9
w • * _1 **
»|»A
: *»• i.s
!«.!*.
IP i
111. -7
IZZ.S
ill. i
\Z4.L
17.1. S
i.s
1.5
13.
6.004- ,
J4
31
104-
O.IS1- , 85°
0. S02._
•l'°_
I 9 2.Z.
, 29.
54-4-
' 544
• 2-4- 544-
(JOOQ4AO)
0.004
O.OT%1
-O^fe-
O.OZ.74- J 0.3fcfo_
o.os \ i
1.0 SI
1.44-1
"I
1.40S
151 .
-------
DAT t. I /ib/74-
PLOtU
DE-VMC.E.S.
2-v«tU ASivXfc. FLOU)
AJOO DRV
tiAS-METtR.-.-
TO
WE.
fxsrv^t. i-
<— V
«u
StCI •
1 — 1
i
i* -
• - c.
•L™^-
? ™^
&AS
fnti tK
L
-rmt PAlCV OF ^vsT&«M"'rb
JN)_
. \3
Mo.
I *
C- 1 -V..-^ i
10
_30
70
__o._i___; iip.j , 0.4
J z!iJk 1 .lit. -
1.4
2.1
3 , O.OOS , 64
. 6
8.5
(po
.M_J_J!2
84. it
0-t- i 12.
'
St I 2.2.
14.5
U I ^-9
\2l.t
0.4
I4.S
31
G(
103
o.oo4- 64
0.507 &4- ! 2.7
.107 ! 8 4 i
I.J44-
64- 9?
C, >vo
1*2-0 >"«-
! ^i
* F
; ^ T».
cum
R.
to
lo
so
70
9o
*>,
M.M
30.05
11.34
33.5%
3U.70
*-~
29. IL
29-25
1} 43
19.75
iO-20
A*
0.00*5
0. 1 4-1
O.SOl
1. lOt
».934
>.
2-9. 2«
i>.?0
31.08
53.2S
2, Co. 13
1
S
0-OTft5
O.ltl)
Q.SCFKW?! 70'F^»«-\«
U.oczc.)
o.ot
0.iS4
0. C.%2
>.049
1.4-SCJ
Q D&IM Cc*»*.)
O.OBbO
O.S5&
o.bbr
0.994-
1.400
1.64*
-------
DATt- 1 /*'/-• r
of
OEVIC.E.S.
Z-v-eU. ASiVlfc. FLOU)
ASfVVE. F
uow SECT. r- *• 1
1
\ 1
1
w-
'^- c.
J™.
TBtT &K
op
BE ComQEtTEO IN) P« W^r.
CALlB.
METE-E. Mo.
a.
to
30
so
10
65
U^)
o.z
o.b
1.4
1.1
^.-^
^,~e
u«o
»»3.3
iil.1
112.0
iiVi
iZS.S
eo«o
^
«.»
t.S
<^.l
B.S
>1.5
1
•p
TI
•2..S
\T>
12
04
100
**
0.007
O.l«»4-
0.41L
I.OfeS"
i.liz.
t
do
90
BO
do
60
1,
I—2- j
to
2*_- j
bo
H
K4^)
«.*
. 0.8
1.5
i.l
Z..7
•V.-"«
u»o
ii^-7
lU.B
I2-7.&
(2.Z.3
izs.s
T^
O.-L
I.S
4.Z
&.(.
n.s
r^oAJ 2.
"P.
•b
»i
34-
t7
too
A|p , -t
o.oot, do
0.152 , 50
o.sit , 80
1,011 : 80
1
1.1,24. 1 ftO
^
16
"
30
bz
94
f^ •
a.
O"
C- ~
UO__iV
?o°F
Q D&WV U*»*-)
40
15.54
b.OOb?
Q OCX.
30.^0
_?o_
J»ft
65
K.
. 97
io.J.3
o.
_°: °yjL
0. 3fc3
.M.9_7_
I.O(oB
13.63
S«VO
540
O.OS14-
0.
O.fc77
0
o 099 8
043
i.oio
035
153
-------
DATt 1/21/74-
OE PL-OtU MfcAS.QfclAJG- OFyyiC.E.1. C
FlLT.J
2-v'«ek ASrwt. FLOW)
3
£
AJOD DRV
vGAS-f>ET-fcR.-r-
TO
• Q «• 3 ABS
ASWt P
J
Mini
1 t "|
1
..U-
" ' l "
| 1
i
i g« .
• - c.
— IL,
| y«.J*t,«L
UK.V
-rmt PAIZ.TOF ^VSTB^ TO
oL Prt fc fciC.
t 29.40
METE-E. Mo. 4-
R.
10
30
SO
10
as
Ui*)
o.t
O.B
>.s
i.i
2.1
•ki-i
U")
liO.O
1x9.3
113.3
1 2.4.0
»39-S
^
O.I
o.C,
l.S
3.1
S.O
T, : A|f» | t
3 , 0.006 < 60
1
iz i 0.11,3 1 Bo j
....
50 0.41) BO
fei \.ozs ! Bo
95 I.S15 80
^
1.8
•
10
2-7
s&
85
: c-
H.UM
L 0.1
r ' " •
, O.B
1.4
1.1
*-7
•\.« «••«
u«o
ni.o
I54.S
ilS.L
12.1.2
>35.0
?-
O.I
o.t
I.I
3.2
S.O
*.
3
^
31
bS
9C
A IF i | 1s,
0.008 BO j (.9
0.113 ftO , 10
(MSB 60 ; 17
1.0*3 ' 80 (»0
l.57t i &0 90
Li^c.
>\,o
BY
K-
(«»*.)
\o
-l^J
.so.
BS
yy»tj -fi-.^-j..
J«.-^.! z>-*.* ; 5iI?A
I.t4_.19-S_|
I.4-!?7.'1''43 ! ''Oi4
19.54
540
o.oou
(o.ookvo)
30.14
5*0
JOiOb —
0.3SO
0. 0901
o. 2.SI
31.59
33.74-
I.OLS
540
ojL^.LE.
i. IS3
154
-------
V
-L.-.
-------
wCv.vi,v >w> c»Trtr >-:-VA
._. * - *•»
-o. 09 (,9
I.012C,
I 4.4-009
c-) = U»«x,»|iJ - *Ui
4« / \_^.UL_Sl9 V^*
..fju)l
^ =
—v
• •- ro)
o .
o.
0.
l.49t£
I.550S ;
a s.^sos
,2 o,
0.191J
O.bDZI
\.ioo«r
.^a96
0.<-S32.
.. 1C •= Q.SI3S
.'. M « ZA.92.
0.
156
-------
ofct
-371&-001
4/9/74-
'
157
-------
11-3718-
4/9/74-
INLET Al*
CAll&£AT10A)
fcv)M 1.
, uwcoefc.
0.1
_o.f.
-Si*.
>*.
1.0
i(o2.
0,9m
II. OS
••_»?>*_
0.9 HO
-i]ps>
132,
.0,9*40
P.963J
0.981?
o.9»fc7
0.9B34
9, ST.
•»•.**_
io:80
(9.96
1Kb,
0.9*31
0,98^,
.MW.
Z.S.O&
_35J,_
°_i?834.
0-9B01
1,4-
2.0
z.s
.2.0
3.S
4.0
f'S
5.0
. 3B3^__
.4_oj._.
O.H70
. frO
3 BO.
0.9*30
O.J8S4-
0.9117
0-9
2.9.
o.nes
.434^.
600,
o.n.24
0.9814
4-T.IO
O.N.14-
_7o4.
0.9*11
_p,9.80Z
_0-i"'7.0
_?.-mj.
0.9717
. 2 S
4-0.78
44. 0&
0.970S"
bH.
5B.S4-
fc.O
7.0
0.9717
0_.9_
^43, I o._9 501
5S.&7
r%. 42.
_p_.9BOZ j C»V Ot_
675,
93S,
t
0.9000
0,9770 !
0.9T3?
158
-------
op
-"i n ' 6 - ool
IM
O.I
I.D
2-C.I7
J.I.
1.4-
31. i?
-
l.S
45.
fft.Oft
u.o
7.0
159
-------
APPENDIX H
OPERATING INSTRUCTIONS AND
THEIR DEVELOPMENT
160
-------
II- 37 IB -001
EfcC.
INIT\ALP. PlIMlPS ON - DUMMY FlUTEE. - *ET FUOUJMETE. E. 5 AT TO
. t OPE-R. ATlKJCr _U>/V>D I.T? 0/SL_ «^ ITH_ _EJC.H
O.S AA)O Z.O t* Hou
" o
— .\DLE._ BACK. pftjESjSoet __ %+VQOCD _at_ .ABOUT- .o.s ^joT^tes . i._o oft
._ .(LA UC.O LATE EXttMJST MASL Fl-OMj. =_ M e . ^_ M^ (A' R-) t ..M^_ (W EL. )
PLO-TROM •. Mq=
_ OK. R£AP fry -Cirw EE.C, 6-KAPft 0
M „ w. if
.3.
LE_ OPERATOR'S .Pl6iT_AL,_g.EADOy.T)- -
rvr"r"i.si
4,_SET_ OIMJTE S/S MPLE FLQIAJ R-ATE_ .AT.. SO _ ACf-M
J^ AT T«is .(.®-°'V-TUWWEL- D16>TAL-
"~ " ' Tt*** .1
S. CALCULATE PgJVtARV D]JLL>T» OA)_ A|». ? 4_lPdx = 0-.*>4MO'fc) y Me O&
3 *""
fe. SET AtV AMD At»Aa.. .0_&lA>& SHUTTER. AMO SILUTiOAJI AlR. tQAJTR-OU
E-R.(L,
.SA_IV»JP ut_JPOJWp. AMD IMS
. .9..
;_ >» MULT A^E-QU SLY . STAILT \ tFM 2>Af« P UE PUIV\P A/OD
RtSET Mi. S R-OiO«WETE.R. [lACFMl . AWD M.. S CAU&eATlOU
- - — '• -'- " --- ~~ — --- — ---- X — * <|^ --
AMD . R.tcoitD DATA.. AT IMDICATEO \AjTEe.v A«-S_OM SAMPLER. OP'S..
|\ tf\V)T DOU)M i C^»M PorVt(> AMD Tl*MEC. (OttEO iWblCATEP OR. if- FILTER PL.OG-S
161
-------
=T==3
-t
fn
II
TO
fill
567
i
nr?
91
It
«ti
78
s
fcti
m7
'TT
Hi t
7t
•:TT.
T •
f
1
tr
t
-• -
Ttn
ft l
- T1
II
rr
HI
7—
I
• •
1 • ,
,
i
if
;i
"tf
i
rrs
i
Uri
t;r!
i
;:;.
1
ttT
ffi
f
.
"T
•«
1
JL
I
ILll
-^
•Ml
T
M
.
h
!l
H
I
I;
11:
Ira
I
!
Si
It1
-i
ft.
O)
1
4 •
t
bii
ill:
H !
ERG
\Ph
•
Ml
1
s
Hi!
i
'
;::|
4^
rrn
-.--.
*
i
«^
T
M
l-f
I I
i • •
s
°L
:Z
I
7891
1
t-H
lit'
• •
^
6789
5678
-------
Mt LOGARITHMIC 46 732O
C, 1X3 CTCLCS ».oi i. u > •
KCUFFCL « IISCPI CO
6 7891
CTi
U)
-------
30 40 50 60
DILUTED SAMPLE FLOW RATE,
•Figure 15. Diagram of the Diluted Sample
Flow Rate
164
-------
T
ER C
GRAPH 3
a 4 5 e 7 * s ic
165
-------
O.lfc
O.I4
i
i
o.\o
O.OB
0.06
0,04
O.02
E R C
GRAPH 4
20 30
4O SO
ft.
feO -7O
166
-------
! " r i "1
r—
zo ^o to eo
_.fH_Ou)METER. R.E.ADIKJO-, % FULL
IOO
167
-------
R-ATE OA-TA.
PR.OBE
LOC.ATOAJ
v\i -v/OL vft *'0 )
4 * 41 tUMAA
EfcC *H*fc/l
No.
»
Z
3
4
e
3
4-
5
l>
Tl P
X.D,ln
o.soo
o.soo
o.soi
o.goi
O.ZS3S
O.iOOff
.o.-ioss-.
O.ISIO
0. IOCS
lAiteA,^1-
9.4U
o.i9(.
o.\9t
o.»n
o.\57
o.osos
o.o^ik
O.OTZ3
0.0119
0.00191
A<-EA , ft*
O.OfcS?
o.ooist
0. OOI5C.
0.001^7
0.00*3?
O.OOQiSO
0.0002-'9
o.ooosoz
O. 000 124-
O.OOOOSS"!
H-V7I6-OOI
7/ib/-74-
W
tVL
lso
I- M
PfcO&E. Nl«.
\
2
S
4-
-HP T.D. ,.-^
o.soo
O.ST>O
o. so (
o.so\
U)
0 00>S\
0 0015 1
0 . 0 0 1 £7_
0 OOIS2
168
-------
II -37^ 6-00
-7/18/74-
RATE =• M C°/A ) E.xH.
ISOKKUETNL
- 02.89
Klo
2
3
4-
£-
G
o-zoos
O.iSIO
O.IOOS
O.OOZ^T
0.0016(0
O.OOIQb
FLOU)
T =
_ A-B42.
'
: M - M .. + M .
/- e. *- T
<- o.s: % «
«t > V. for
DR.Y
1.044
i-o^o
1.0
169
-------
-vn B-OO \
7/1 7/14
*T * ^f-^J7.^ ?P _.y.EN."!V.^ CUJ-VEC-E_ec
_ __ __
It. ll&) V. *•
Q i. II ___ \.14 1.-9 ____ 3.4t_ 5.89
" " -
.
___ ^ ____ Q_.OT70_ _J>.
OAtA_ TV^^OO^ Pxs.^.^ f* ^y '-. {.=• o.
4- fc 10
" ---- ---
0-071%
8 »0
(4.5B8;
_____ o.oi 1J __ O.JJ.S __ Q.H.1
ALL.
100- Tv «= 300
1.0 -
>- - *b~- = °'*Bb— =-
11 TO">
-------
•7/I&/14
Of A)>d »N TtR-lw£. OF- VCNOVOfO Q
_t ^
•. tb^ «JU
V- 7Y*__,1,_ A>^-o.«M4j^J_.j_QA^_LA.^
M^
:fcr
kft.
>l
/
... I- /J> ^V*
n«i
F-OH
Cl*
cJOL /
NOTE: T us&o
-------
NOTE. T& USED
Tv ON
DATA
11-3716-001
•7/19/74-
Ot/OtlUVTE.
M
AS PARAMETER.'!.
Tte.«.lOO
Ts
eio
s-to
(b \ 0
(i> (B 0
1 \ 0
IfeO
*v=L-«<'o-)T^vr* = *(*>
K ^|IhT.,A ^4,.,
l.UUl o.z^ff l-fcfci
I.6B2 O.bl4- t.%%2-
z-l 1 O 0- ^75_ z.i lo
2.W3 o.4lt> *••*«
2.583 0.4-SJ l.Sll
Z.6Z8 0-SOl 1,828
•v)1''1*
^1^=3
S.4I
t.12.
b. 87
7.G1
8.41
9.2£
srio
S<» O
(b I 0
fcbO
HO
7fcO
t.bSI
l.ftll
2.09(0
2-i29
4-Sfc7
2.8 11
SIO
SfeO
G> D
(oUO
110
2.064-
2 ^1 4-
Z.S5I
2.T93
172
-------
W TV©OW ONTA --o
"7/19/14-
4(L-
R.
_.. _.. ii
* U-IOKio-') T/ (^v)
(olD
fcfcO
1 10
— - ^f ^'S - -.__..._...
i i 9_
NQ
SO- «JSE FlC-OrtES .POR. TO F (^tu»TKf»J \% (oO F > Td^ t 80 F ) A^ D
i^.-i I* W. ^IO»THI>J I0/. ^9.0 £= jf*.. iz Z^-C )
173
-------
1 TxQy ON OftVA \\-3ll6-OCM
SHEETS.
y E.SU F-\C/VTi(yO^ OF &>„ vt Afc^ g.ELAno/0 SH_»_P_
*'6 ''
Afc. = I.SB ^. rt.O VEgtFl
•* aft. —— *" — -•• •-
70F
174
-------
G-EAJEI2.M- (LOW
z.9.0 ^
/>cu»r /M^loAi tv— W^
-------
\\-Vl»&-00\
OPE.K.ATIN>G- INJSTHUC.TIOAJ&
HI -VOL.
FILTER-) V/ iT EM
I. SET &|oo(l (^ORIFICE IN TAILPIPE) = Z.OS I* HtO U«TH
OFF- AAJD CMUOTIOM TON MEL BLOWER OM — ^tOlTCH OFF
2.. START EWfclWt AiND EiTA6UiH OP£tA-T\WG- C_O AJ O iTl 0 Aj
AS POS,SlBLt
2,.
AS
4.
OF
8UXJJEt,COUKnEI?,#t
S»»WLILTAA>&OU^LY AT
B.
SET.
CAM B K-ATl 0*0 _CUI2.\lE,S
FU»w>N\tTEIZ.S
OFF AS ^OL.LO«AJS, L)SI /OC-
rWETEE.?. 1 — 4- (jONMEL BUKUEJt OK ) !
TO
\%
2. STOP PurnPS> AW£> TAKE "BER>R.E* &R.V G-Ai. KlETtK. REAO/NJG-S ,TrHE\)
AWD ESTABLISH COMDlTlOlO AS QOlCK-LY AS
3. STAtT
WfcCJESSAPV OU(JaM&- B4)«NJ OSiMG- VALVES
A.
AT &AJ b 0
176
-------
177 •
-------
APPENDIX I
PROCEDURES FOR DATA REDUCTION
178
-------
FOR MUL.TIMODE
SAMPLER
RUNS
OPERATOR
DATE
RUN
MODE
1
2
3
4
5
6
7
8
9
10
1 1
12
13
TIMEC*«ejAT
START
END
ENGINE
RPM
UOAD,
Vb
TEMP'S., *F
TV (D
TV©
To*©
To.®
Ap, Ln HjO
RRIM.
DtW.
MAW ••
IVIMTUIII]
OIL.
SAMPLL
FLOWMfcTER R.
1
1
3
4
5
ORIFICE
Ap
FOR SINGLE-MODE RUNS (.STEADY-STATE)
1
z
I
4
S
6
7
8
9
ID
II
U
U
TIME.
see
TIME,
mln
TEttP'S., *F
TP©
T^©
Ti»(3)
T.»©
Ak in H,O
PRIM.
Oil..
MAW «.
MNTVII)
PIL.
1AMPLE
FLOWMETER R.
1
2
3
4
5
ORIFICE
Ap
1
2
3
4
s
FILTER No.
BY POSITION
1
2
3
4
5
R
FP47-
A47-
AA7-
A41-
FP47-
AR.-
DRY GAS METER
ftEPORE
APTER
ENOINE
ERC PROBE T»P i.O.
-------
TUNNEL.
FOR MUUTIMODE RONS
OPERATOR
DATE / /
RUN
MODE
I
2
3
4
5
6
7
&
9
1 0
i 1
I Z
1 3
TIME (MO AT
START
END
EN&INE
RPM
v£°'
TfcMP'S., »F
Tk®
T,®
T;©
TudD
T«©
BieiOER
A|P
?>»
VAiyETOftKSCUl
UAR&E
SMALL
NO, RAW
RAAJ&E
CHART
NO. OU.OT6
RAMfrE
^
CHART
o FO
RP
I
1
a
4
S
6
7
8
»
•
u
u
R. &IN&UE - MODE RONS tSTHAOV - bTAT E)
K = Tt UO AD « LOAD Clfci) B
TIME.
sec
TIME.
WIIN
TEMP'S.. %P
TA(§)
Tkd>
TF©
TcmOD
TL©
BtOWEK
&l»
^*
MLVETURN&CLO.
LAR&E
SMALL
TUNNEL BLOWER COUNTS | t
SANIPL1N& Tti*\t. *«c t m J
NO* RAW
RAN&E
CMART
Wt DILUTE
RAMGt
ttVMIT
-------
FOR. MOLTIMOOE RUNS
MODE
1
2
i
4.
S
-------
U-37»6-OOV
6/13/74.
DATA
DATA
ONE. VAL.UE. PER. TS ST OtjE-VACUE. P££ MOO
_ ____ ___ ....... __ . . **»y'»e air
blew**- *H»^t »H ta»»i * /.*_«€.." 1^ r.._KCV.l!"L) ____________ tfcji-n «. ctl i> i--w^« r^i\u^i •«• Tt f "F
_^) ______ _______ ____ W»*«-r
i K V . ^^ ____
N . ____ _______________ Kc-vfll efface *> «• A9 , 1*
V**1
flk-/ "S
. _V._ _7»»*n.J_
E._,_...._.^jio. v
FOR m-VOL ^At>APLBS ONL.V
i
iSl
"^
182
-------
tl-Vl Ifc-ool
~~ (.TON/MEL) CAMPLES. 0/VH-V
s O..OTSS
*M N
V.-I
41 ~^
./ x -
tiwCitioK /•ft'Vl L 'Kr / ~
183.,
-------
PAGE 1 OF 2
1 2
>u
RUN
NO.
d-
-
d
d
5
d
678 9|ldlljl2
laddie
»EQ DATE
mm/ d d
oil /I 1
FUEL
CODE
did d
0(2 |E M j
/
/
V
V
G
d
F
17
IE
19
20
RPM L(
d
&F
d
!
V31
dd d' d
03| i
•
0 4JT 1 M E|
i
0 5] lT|OiR[
oe| | T|E|
i
i
d
1° 7I T|BI 1
d
d
d d
d
\ OR G47
d
d d d
V21
|
d
d
d
^
d
1
d
2
d
~|~
1
d
—
d
1
d
2
o
d
1 •
^
2
i M
21 22!23!24 25{2S27J28 29J3C
AD >
NO 1 - R
d d a a did -d d d
i _ 1
1 i
Pt
313^33)3435
N(
d
1
.RTICULAT
1 G472 G473
d ;
I i I
DRY
V12
i
1 ' 1
!
3 4
1
1 ' 1
3 !4
i
1 i 1
i
3 "4
!
"1 1 i |
i B
3 |4
1
GAS METE
V22
TIME
5
6
ORIFICE 1
5
6
36J37J36
39
40
). 2
-
E
I
WEIGHT
R
G
READIN
V13
N
MODE
7
EMP. BY
IV
7
NTAKE Al
5
6
LOWER IN
5
6
i ' 1 , • ' 1 1
<*TEMP
7
414243l
-------
PAGE—2 OF —L
1 2
3
4
5
RUN
NO.
d-
-
d
d
_
d
-
678
SEQ
9
0 B|D'P
09|P
U
1.|
D
1 1JD
N
1,)*
N
10111
1
i
i
OJRJ
E
P
0
O
L
*
x
12
d
d
X|
^
14
IE
1
d .
d
1
d d
1
d d
1
d d
1
d
d d
T
16
d
17
18
19
20
2
2
d
2
d
2
d
2
d
'
-
2l|22
23
24
25
3
3
3
26
27
28
29
4
30
5
4
1
3
I
T
3
|
|
4
5
31J32 3
33435
ORIFICE A
6
1
36373839
40
» FOR HI-V
7
I 1 '
ENGII
6
1
OE FUEL
R
7
j |
ENGINE,
Dl
4
1
F
4
r
f
5
1
MRFLOWC
6
~ri
LUTE NOxiCONCENT
5 6
1
AW NOX C
5
|
_L
11
42
43
44
45
OL SAMPL
8
II
ME
IN LB,
8
RIFICE&P
T
~|
NATION IN
7
1
DNCENTRATION
6
IN
P
7
i
m ~
1
46
4T|48 49 50
E BY MODE
9 1
,,/HR BY M
J
) 1
~1 1"""
BY MODE
8
51
K
5;
54
5E
) 11
DDE
3
11
9
_
PPM BY M(bDE
8
'M BY MOC
8
1
9
E
9
|
5f
5:
59,60|61J62J63JB4
I
12 • 13
L
, I ' : !
65
6E
67
68
S9|70
12 , 13
i
iM: i
10 ! 11
!
: • 1 1 1
10 j 11
!
! ' 1 1 I
i
10 ' 11
1
H-TTTT
_
12
12
.
1
12
P
»
73
74
75
13
, 13
13
1
76)77|78 79)80
1
1
1
-------
000003
000003
000003
000003
000003
000003
OU0003
000005
000012
000021
00002H
OOQ02b
0000*7
000052
OOOQSb
000112
000132
00013*
00013S
OU013b
0001*0
0001*1
0001*2
0001*3
OOQl*b
000151
000153
00015*
00(1155
000173
000177
000212
00021*
000220
000230
000257
0002b7
000271
00030H
00031*
00032*
OU03*0
OOJ3*1
000355
PROGRAM *ATECC( INPUT, OUTPUT, TAPEbO=INPuT)
DIMENSION TORC13),DPE(13),TE(13),FUEL(13),RNOX(13),DNOX(13)
DIMENSION TB(13),OPOR(13)»TIME(13)
DIMENSION EXHU3),AIR(13),FLO*OR(13)
OIMENSIOM SAMP(*),ENGCOOE(2)rIFLTRS(5),GJ(S)
EQUIVALENCE (GJ(i) , G* 7l) » (GJCg) ,G* 7g) » (GJ(3) , G» 73) i (GJ(«f )
EfJUIVALENCE (GJ(5),GERC)
INITIALIZE ARRAYS TO ZERO
301 00 1 1=1,13
TOR(I)=OPE(I)=TE(I)sFUEL(I)=0
R*OX(I)rDNOX(I)=TB(I)=OPOR(I)sTIMECI)=0
EXH(I)=AIR(I)=FLOWOR(I)=0
1 CONTINUE
READ HEADER CARD
READ(bO,lUO)JRUN,JSEQ,JDATE,JRPM,LOAD,IFLTRS,ENGCODE
IF(EOF,bO) 80, 2
2 PRINT 200
RE AD ( faO, 1U1 )KRUNfKSEQ,FuELC,G* 71,0472,6*73, G>»7*,GERC, PA, CB,TOTlME
1 rN
PRINT 201,JRUNrJDATE,ENGCODE,FUELC,LOAD,JRPM
CHECK FOR FILTERS
IS=1HS
IERC=Q
IFCT=1
IFCIFLTRSC2).NE,9H ) GO TO 3
IS = 1H
GO TO S
3 IFCT=2
IFCIFLTRS(3).NE.SH
)IFCT=3
)IFCT=*
IF(IFLTRS(S).EQ.SH )GO TO 5
IERC=1
IFCT=5
5 PRINT 202,IS,CL,IFLTRS(L),L=1,IFCT)
C IF IS=BLANK THEN THIS IS A 8X10
C SAMPLING SYSTEM.
C IF I£RC=1 THEN THIS IS AN ERC
C SYSTEM.
C THERE SHOULD BE A PARTICULATE
C VALUE FOR EACH FILTER SO LETS
C GO.
IFCIS.EQ.1H )GAR=G*71
PRINT 203,CGJ(L),L=1,IFCT)
IFCIS.EQ.1H ) GO TO b
PRINT 20*
b IF(KRUN.NE.JRUN.OR.KSEQ.NE,2 )GO TO 8
READCbO,lU2)KRUNrKSEQ,VllrV21,V12,V22,V13,V23,Vl*,V2«f
IFCKRUN.NE.JRUN.OR.KSEQ.NE.3 )GO TO 8
IFCIS.EQ.1H ) GO TO 30b
PRINT 205,Vll,V21,TOTIMErPA
PRINT 20b,Vl2,V22
PRINT 207,V13,V23
PRINT 208,Vl*,V2*,CB,N
GO TO 7
30b PRINT 218,TOTIME,PA,CB,N
C READ REST OF DATA FOR ALL MODES
7 READfbO,lU3)KRUN,KSEQ,CTIME(I),I=l,N)
186
-------
000371
000101
000122
000*32
000150
OOOIbO
00017b
OUOSOb
000521
000531
000552
0005fa2
OOOfaOO
OOOblO
QOQb2b
oonb3b
OOObSI
OUObbl
000b?0
0007CO
000702
00070b
000735
00073b
000712
000775
000777
OOJQOJL
00.1011
001015
0111017
Q01Q21
OOJQ22
001023
OUJ021
001031
001037
OOJ011
00-'Gib
001053
001051
OOlOSb
001057
00l0b2
OU10b3
OOlObS
0> 1070
001071
OUJ072
C
C
C
C
IF(KRUN.NE.JRUN.OR.KSEQ.NE.I) 60 TO 8
REAO(bO,lll1)KRUM,KSEQ» CTO«(I) , I = 1,N)
IF(KRUN.NE.JRUN.OR.KSEQ.NE.S) GO TO 8
REAO(ba,luS)KRUN,K3EQ,(TE(I),I=l,N)
IF(KRUN.NE.JRUN.OR.KSEQ.NE.b) GO TO 8
READ(bO,lUb)KRUNrK3EQ,CTB(I)rI=l,N)
IF(KRUN.NE.JRUN.OR.K3EQ.NE.7) GO TO 8
READ(bO,lll7)KRUN,K3EQ,(DPOR(I),I=l,N)
IFCKRUN.NE.JRUN.OR.KSEG.NE.S) GO TO s
READ(bO,lU8}KRUN,K3EO»(FUEL(I>»Isl»N)
IF(KRUN..ME.JRUN.OR.KSEQ.NE.1) GO TO 8
READCbO,lOcUKRUN,KSEQ,(DPE(I),I=l,N)
IF(KRUN.NE.JRUN.OR.KSEQ.NE.10)GO TO 8
READ(bO,110)KRUN,K3EQ,(DNOX(I)f I=1,N)
IF(KRUN.NE.JRUN.OR.KSEQ.NE.11)GO TO 8
RtAD(bo,lll)KRUN,K3EQ,(RNOX(I),I=l,N)
IFCKRUN.NE.JRUN. OR. KSEQ.NE. 12)60 TO 8
IF IS=S THEN TOR AND DPOR ARE
BLANK
IF(IS.EQ.IHS) 9,10
8 PRIM 20S,KRUN,KSE<3
STOP 10
»X*7 SAMPLE OUTPUT FORMAT
9 PRINT 210
PRINT ail,CTIME(I),DPE(I),TE(I>,FUELCI)rRNOXCI),DNOX(I),TB(I)
12
GO TO 11
10 PHINT 212
PRINT 21?,(TIME(I),TORCl),DPECI),TECn,FUEL(I),RNOXCl),DNOX(I)
1 rT3Cl),OPOR(I),I=l,N)
CALCULATIONS COMMON TO ALL
SAMPLE TYPES
11 R=C8/TOTIME
00 12 1=1,N
AIR(I)=102.7*(DPE(I)*PA/(TE(I)+1bO,))**.*812
EXH(I)=AIR(I)+FUEL(I)/bO.
CONTINUE
CALCULATIONS FOR HI-VOL
IF(IS.EQ.lHS) GO TO 11
SMEXTI=0.
SMFLTNsn.
00 13 1=1,N
FLOWOR(I)=5.35*(OPOR(I)*PA/(TOR(I)+1bO.))**.5
3MEXTI=SMEXTIfEXH(I)*TIME(I)
SMFLTN=SMFLTN+FLOWOR(I)*TIME(I)*DNOX(I)/RNOX(I)
CONTINUE
PER= 3bOU.*GAR*SMEXTI/TOTIME/3MFLTN
PRM= PER/GAR
PCM=PC/GAR
PRINT 213
GO TO IS
ERC SYSTEM
11 IFCIERC.NE.1)GO TO lb
PER=eJ.b8£ + 7*GERC/TOTlME
PRM=PER/GERC
SUMEXH=0.
00 15 I=lfN
13
187
-------
0(11073
001075
001077
001101
001105
001110
001111
001113
001117
001121
001123
00112b
001130
001132
001135
00113?
ooim
opim
ooim
0011*7
001150
001151
00115*
OOllbO
0011b2
OOllbl
001170
001172
001175
001177
001203
001205
001221
Q01223
001221
00124Q
0012*1
001212
0012*1
0012*1
001211
00121*
001211
001211
0012*1
OOJ2*1
Q0121*
0012**
001211
00121*
00121H
001211
OOJ2H
17
SUMEXH=SUMEXH+EXH(I)
is CONTINUE
PC=l.2b3E*12*GERC*N/TOT!HE/8UMEXH
PCM=PC/GERC
PWINT 21*
60 TO 19
1 X 17 SYSTEM
lb V=V2l-Vll
IF(V.LT.O)V=V+100.
SAMP(l)s.u753*V
V=V22-V12
IFCV.LT.O)V=V+100.
3AMP(2)=.U7b7*V
V=V23-V13
IF(V.LT.O)V=V+100.
SAMP(3)=.U7S9*V
V=V21-V1»
IFCV.LT.O)V=V*100.
3AMP(4)=.U?55*V
SMEXTIsQ.
SMTINO=0.
00 17 I=1,N
3hEXTI=SMEXTI+EXH(I)*TIME(I)
SMTINOsSMTINO+TIME(I)*ONOX(I)/RNOX(I)
CONTINUE
RATIO=SMEXTI/SMTINO * bO.
PRINT 215
DO 18 J=l,1
PER= 6J(J)*RATIO/3AMP(J)
PRM= PER/GJCJ)
PC= 2.b13E+b*GJ(J)*TOTlME/SMTlNO/SAMP(J)
PCM=PC/GJ(J)
PWINT 2lb,PER,PRM,PCfPCM
CONTINUE
GO TO 20
19 PRINT 2lfafPER,PRM,PCfPCM
IFCG172.NE. 0) GO TO lb
20 GO TO 301
80 STOP 100
INPUT FORMATS
100 FORPAT(A5,I2,A8,A1,A3,5A9,A10,A3)
101 FORMAT(A5,I2,Ab,SF8.b,F5.2fFb.O,Fb.l,8X,l2)
102 FORMAT(AS,I2,8F7.3)
103 FORMAT(AS,I2,SX,13F1.0)
101 FORMAT(A5,I2,5X,13F1.0)
105 FORr"AT(A5,l2»SX,13Fl.o>
lOb FORMAT(A5,l2f5X,13F*.0)
107 FORHAT(AS,I2,SX,13F1.2)
108 FORMAT(A5,l2f5X,13F*.l)
109 FORMAT(AS,I2,*X,13FS.2)
110 FORrtAT(AS,!2,1X,13FS.l)
111 FORr*AT(AS,l2»1X,13FS.O)
18
200 FORMAT(*1 TABLE
1TRATION CALCULATIONS* )
201 FORMAT(30X,*RUN *,A5,2X,A8
1 * LOAD *,A3,* RPM *,A1 )
202 FORMAT(*U FILTER*,Alf12,IX,A9,1(13,IX,A9))
OUTPUT FORMATS
PARTICIPATE EMISSION RATE AND CONCEN
FUEL *rAb,
,188
-------
OU12HH
0012HH
0012*1
0012ft
0012*4
0012**
0012**
0012**
0012**
0012**
0012**
2133 FOR'
-------
TABLE PANICULATE EMISSION RATE AND CONCENTRATION CALCULATIONS
RUN 2-000 08/21/7*
ENGINE O.D.bL71-T FUEL E«198F LOAD RPM COMP
FILTERS 1 FP*7-lb8
PARTICLES .000929
DRY
* X
2 A*7-179
.001838
GAS HETER READINGS INITIAL
TIME DPE
198 g.rjo
183 2.30
208 2.70
222 3.*0
2*0 *.30
182 .15
235 8.*0
238 fe.BQ
23* 5.50
21b *.SO
219 3.80
*7 SYSTEM
PARTICULATE
PARTICULATE
PARTICULATE
PARTICULATE
PARTICULATE
PARTICULATE
PARTICULATE
PARTICULATE
1 8.b3Q
2 *3.b*3
3 71,358
* 58.338
TE FUEL
fa? 11.0
b? 23.0
b9 *S.O
b8 bB.O
b9 85.0
70 1.5
71 10*. 0
71 78.0
72 59. 0
70 **,0
70 19.0
EMISSION RATE
MULTIPLIER
CONCENTRATION
MULTIPLIER
EMISSION RATE
MULTIPLIER
CONCENTRATION
MULTIPLIER
EMISSION RATE
MULTIPLIER
CONCENTRATION
MULTIPLIER
EMISSION RATE
MULTIPLIER
CONCENTRATION
MULTIPLIER
3 A»7-180 *
.001873
FINAL TOTAL
A*?-181
.001911
TIME ATM
lb.157 237*. 3
b2.8*0
90.827 BLOWER
78,*32
RNOX DNOX
785,0 3.9
2*0,0 11.2
503.0 2*.?
855.0 **.b
1152.0 b3.5
191,0 5.9
1172,0 93.5
883,0 b3.5
5*1.0 3*. 8
273.0 lb.3
9bS,Q b.l
= 99,18
= 10b7b2
= 89157
= 95970825
a 75.5*
= *109b
= b?900
= 3b9*259*
s ?b.70
s *09*9
= b89*b
= 3b8lO*l*
a 7b.22
= 3988b
= b8518
= 3585**2b
COUNT NO.
b*80!
TB
75
75
80
8b
95
90
9*
109
IQb
99
93
PRESSURE
29. *8
OF MOOES
11
190
-------
TABLE PARTICULATE EMISSION RATE AND CONCENTRATION CALCULATIONS
RUM 3-000 08/13/71
ENGINE D.D.bL71-T FUEL EM118F LOAD RPM COUP
FILTER 1 AR-73
PARTICLES .097900
TOTAL TIME
2375.5
ATM PRESSURE
21.15
BLOWER COUNT
b4lfa8
NO, OF MODES
11
TIME
118
183
208
222
240
182
235
238
234
21b
211
TOR
7b
84
88
12
1b
11
101
lOb
108
107
105
DPE
1.85
2.00
2.30
3.20
3.10
.14
7.50
b.SO
5.20
4.30
3.bO
TE
72
72
72
73
7b
7b
7b
77
78
71
77
FUEL
4.0
22.0
40.0
bS.O
81.0
2.0
102.0
80.0
bl.O
28.0
20.0
RNOX
80,0
235.0
415.0
830.0
1050.0
220.0
1037,0
825.0
520.0
2b2vO
100.0
DNOX
3.5
10.3
22.5
42.0
58,0
b.2
80.0
51.0
31.0
14.5
5.2
TB
71
81
85
10
11
92
105
110
107
101
15
OPOR
2.05
2.05
2.05
2.05
2.05
2.05
2.05
2.05
2.05
2.05
2.05
HI-VOL
PARTICULATE EMISSION RATE = ?3.i4
MULTIPLIER = 747
PARTICULATE CONCENTRATION = b8832
MULTIPLIER = 703084
191
-------
Appendix B3,;ll
SOUTHWEST RESEARCH INSTITUTE
8500 CULEBRA ROAD • POST OFFICE DRAWER 28510 • SAN ANTONIO TEXAS 78284
September 10, 1974
TO: Dr. Ronald Bradow, Project Officer
Environmental Protection Agency
Research Triangle Park, North Carolina 27711
FROM: Charles T. Hare and Karl J. Springer
Department of Emissions Research
Southwest Research Institute
8500 Culebra Road
San Antonio, Texas 78284
SUBJECT: Monthly Progress Report No. 14 for the period August 1
to August 31, 1974; Contract No. 68-02-1230, "Develop-
ment of a Methodology for Determination of the Effects of
Diesel Fuel and Fuel Additives on Particulate Emissions, "
SwRI Project No. 11-3718.
PREPARED
FOR: Environmental Protection Agency
Research Triangle Park, North Carolina 27711
Introduction
The purpose of this project is to develop and demonstrate a test
procedure suitable for characterizing the effects of diesel fuels and fuel
additives on particulate emissions from automotive (truck and bus) diesels.
The scope of this work includes construction and use of a dilution tunnel
for diesel exhaust, use of a variety of particulate sampling devices and
techniques, and application of a wide range of chemical analyses to diesel
particulate. This report covers the test protocol which has been developed
to meet project objectives, including all assumptions and calculation tech-
niques.
Progress During the Period August 1 to August 31, 1974
The initial batches of samples have been analyzed gravimetrically,
and a number of samples have also been analyzed for total organic solubles,
BaP, IR spectrum, S-C-H-N, and metals. Several samples of organic
solubles have also been analyzed for S-C-H-N-O and hydrocarbon boiling
point distribution. This progress report contains these initial analytical
results; and while some "gaps" certainly exist at this point, the data show
some very interesting trends.
SAN ANTONIO. HOUSTON. CORPUS CHRIST). TEXAS AND WASHINGTON, D C
192
-------
Initial Gravimetric Results. Detroit Diesel 6L-71T
A gravimetric determination is being made on every filter run
during this program, with resolution to 1 J{ g of particulate for 47mm
filters and to 0. Img for rectangular (hi-vol) filters. A computer pro-
gram has been set up to calculate total engine emissions in g/hr and
raw exhaust particulate concentrations in-yg/m from the particulate
weights and operating variables. A number of samples have been pro-
cessed by computer, mostly those collected on hi-vol filters, and average
results available at this point are summarized in Table 1. Note that
concentrations are expressed in mg/m* rather than^g/m^ to make the
numbers easier to handle. For the few runs already processed using
47mm filters, agreement with the hi-vol results seems reasonably good.
Initials. C. H. N, and O Results. Detroit Diesel 6L-71T
Quantitative determinations of S, C, H, and N on 20 sets of filters
were conducted by Galbraith Laboratories, along with quantitative S, C, H,
N and O on 4 samples of organic solubles. The results exhibited quite a
bit of variability, some of which was expected due to the range of operating
conditions and fuels used. The values submitted by Galbraith are presented
in Tables 2 and 3. Although no detailed analysis has been conducted, it
appears that the particulate samples from runs on fuel B exhibit higher
sulfur content than those from runs on fuel C. It also appears that the par-
ticulate carbon content is higher for fuel C than for fuel 6.
Initial Results on Total Organic Solubles and BaP
As at the end of the reporting period, total organic solubles and
BaP have been determined for 10 samples. Results of repeatability
checks and "spiked" filter runs are not available now, but they should
be by the end of September. Organic solubles /BaP results are given in
Table 4, in addition to emission rates and raw exhaust concentrations of
BaP calculated from (tentative) operating data. Variability in the BaP
data is rather high, and it is hoped that some of the "checks" noted above
will point out some of the sources of variation.
Initial Results of Analysis for Particulate Metal Content
Rather than re-tabulate the data on metals found in 14 samples by
Scanning Electron Analysis Laboratories, their data as submitted have
been reproduced and are included as Appendix page A-2. Thus far, no
functional relationship between metal content and operating condition has
been discovered, but more data should help to illuminate this area. With
the exceptions of lead in two samples and an (apparently fluke) amount of
Barium in one sample, Zinc is the only element found in substantial
quantity. It is suspected that this material may originate in the lubricating
oil, and procedures are being instituted to check on this suspicion. The
193 .
-------
ID
TABLE 1. INITIAL AVERAGE* GRAVIMETRIC RESULTS, HI-VOL (GLASS FIBER)
FILTERS USED FOR TESTS ON DETROIT DIESEL 6L-71T
Fuel B Fuel C Fuel A Fuel A + DII-2
Operating
1600
1600
1600
1600
1600
2100
2100
2100
2100
2100
Idle
rpm
rpm
rpm
rpm
rpm
rpm
rpm
rpm
rpm
rpm
Condition
- 0% load
- 25%
- 50%
- 75%
load
load
load
- 100% load
- 0% load
- 25%
- 50%
- 75%
load
load
load
- 100% load
g/hr
9.16
42.2
48.
83.**
87.
102.
63.4
69.
114.
110.
144.
mg/
45
55
62
95
88
89
59
61
89
80
90
m3
.6
.0
•
. **
•
.0
.6
•
.8
•
.4
g/hr
7.8**
41.6
62.
76.2
78.
89.
72.
34.
110.
120.
mg/m3
36.**
54.3
74.
84.4
77.
77.
67.
29.
89.
85.
--
g/hr
3.9
_ _ _
59.
---
. _ _
mg/m3
21.
65.
._.
_ _ _
g/hr mg/m3
4.3 20.
_ _ _ —
62. 67.
___
_ _ _ _ _
Composite 86.7 79.0 76.4 71.9 82. 76. 76.1 69.4
*points representing only one run contain 2 significant figures
**unacceptable variability, 2 significant figures shown
-------
TABLE 2. INITIAL SINGLE VALUES FOR S, C, H, N, AND O
IN ORGANIC SOLUBLES SAMPLES BY OPERATING CONDITION,
DETROIT DIESEL 6L-71T ENGINE AND FUEL B
Percent of Species by Weight
in Organic Solubles
Operating Condition C H N
Idle 85.5 13.2 0.20 0.32 0.73 100.0
1600 rpm - 50% load 83.3 12.8' 0.36 0.18 3.4 100.0
2100 rpm - 100% load 78.8 11.7 0.70 0.25 8.4 99.8
Composite 82.0 12.4 0.20 0.77 4.5 99.9
TABLE 3. INITIAL SINGLE VALUES FOR S, C, H, AND N
IN PARTICULATE SAMPLES BY OPERATING CONDITION,
DETROIT DIESEL 6L-71T ENGINE
Wt. % - Fuel B „ Wt. % - Fuel C
Operating Condition C H N S L% C H N S
Idle 60.9 7.7 "0.1 3.0 71.6 68.4 9.1 <-0.1 2.1 79.6
1600 rpm - 0% load 69.1 10.3 1.0 2.1 82.5 73.3 11.1 0.1 0.785.2
1600 rpm - 25% load 76.8 12.8 1.0 2.6 93.2
1600 rpm - 50% load 63.5 10.9 *0.1 2.9 77.3 77.2 10.7 0.3 0.6 88.8
1600 rpm - 75% load 71.1 10.5 0.3 2.1 84.0
1600 rpm - 100% load 66.7 8.8 0.8 2.3 78.6 75.0 ?* 0.1 0.5 ?*
2100 rpm - 0% load 66.6 9.8 0.8 1.8 79.0 72.7 9.7 0.9 1.084.3
2100 rpm - 25% load 65.9 10.4 <-0.1 2.8 79.1
2100 rpm - 50% load 65.8 9.8 *-0.1 2.4 78.0 77.6 11.7 0.6 0.6 90.5
2100 rpm - 75% load 69.9 ? * <-0.1 1.6 ?*
2100 rpm - 100% load 70.7 10.9 0.4 2.2 84.0 74.5 12.0 <-0.1 0.987.4
Composite 71.8 10.0 0.3 2.0 84.1 75.0 11.3 <-0.1 1.7 88.0
""indicates questionable data
195
-------
TABLE 4. INITIAL DETERMINATIONS OF TOTAL ORGANIC
SOLUBLES AND BaP, DETROIT DIESEL 6L-71T ENGINE
wt. % BaP in Extract
wt. % BaP in Particulate
Operating Condition Fuel A Fuel B Fuel C Fuel A Fuel B Fuel C
Idle
1600 rpm - 50% load
2100 rpm - 100% load
Composite
0.0247 0.0126 0.0148
0.0188 0.0122 0.0172
0.0127
0.0137 0.0265 0.0088
0.0093 0.0036 0.0063
0.0136 0.0061 0.0124
0.0066
0.0087 0.0117 0.0054
BaP Emissions,
Operating Condition Fuel A Fuel B Fuel C
Idle 365. 299. 256.
1600 rpm - 50% load 7980. 4160. 9120.
2100 rpm - 100% load --- 9880.
Composite 7170. 9600. 3950.
BaP Cone.,
Fuel A
2.00
8.82
___
6.64
Fuel B
1.58
4.79
6.20
9.01
Fuel C
1.44
10.2
.._
3.72
data on metals are expected to be more revealing when fuels with metal-
containing additives are used.
IR Spectra and Paraffin Boiling Point Distribution
Infrared spectra have been run on the same samples for which BaP
data were presented, but the traces have not yet been reduced in size suf-
ficiently for inclusion in this report. The major results appearing so far
seem to be indications of some carbonyl and hydroxyl groups, but solvent
interference is causing problems in some areas of the trace. Efforts
are being made to refine the technique where necessary.
Separation of the paraffin fraction from the remainder of the organic
solubles is going well, and a few samples have been analyzed chromato-
graphically. It appears thus far that most of the paraffins are out of the
normal diesel fuel range (nominal C?6 as compared to nominal Cj6 for
fuel), and work is underway to come up with a tracer compound which
should tell us whether or not the paraffins are related to the lubricating
oil. The major problem with the tracer thus far is that components like
n-tetracontane (C4Q) would cost some $1000 for enough to treat the oil
at a 1% level.
Project Schedule
Despite good faith efforts to conclude the technical effort within the
time allotted under modification no. 1 to the contract (68-02-1230), it now
196
-------
appears that another no-additional-cost time extension will be necessary to
accommodate the test operations. The current date for the end of the
technical effort is October 23, 1974, and the date for end of contract is
February 22, 1975. The memorandum initiating action to request a three-
month extension on the above basis has been submitted, and a copy is
included as page A-3 of the Appendix.
Possible Problems and Corrective Action
The only problem remaining at this time is interferences in the
Phenol analysis, and action is being taken to find a solution. Several
filter treatments and new environmental control for the filters are being
tried, and the results should be available during the next reporting period.
Plans for the Next Reporting Period
It is planned that sample acquisition and analysis will be continued,
and that all gaseous emissions and smoke tests on the Detroit Diesel
6L-71T engine be completed.
Submitted by:
Charles T. Hare Karl J. Spitfnger
Manager, Advanced Technology Manager
Department of Emissions Research Department of Emissions Research
197
-------
RESULTS OF X-RAY FLUORESCENT ANALYSIS
DIESEL EXHAUST PARTICULATE ON FLUOROPORE FILTERS
(in micrograms per square centimeter)*
ELEMENT £aV MnNiCuZnPbSr
U.D.L.** 1.70 0.38 0.49 0.16 0.10 0.12 0.22 0.22
Sample
Identification
S-l FP47-107 -- - - - 3.00 -
S-2 FP47-113 -
S-3 FP47-105 -
S-4 FP47-141 - tr
S-5 FP47-142 -
- S-6 FP47-144 -
CO
S-7 FP47-145*** -
S-8 FP47-170 -
S-9 FP47-171 - tr -
S-10 FP47-172 -
S-ll FP47-174 - tr -
S-12 FP47-175 -
S-13 FP47-166 -
S-14 FP47-168 - tr tr
* Analyses in which counts were obtained for
minimum detectable limit are noted by "tr"
detected. •—
1.94
0.54
0.57
2.49
3.16
0.87
1.29
0.33
2.85
6.09
1.89
2.95
1.95
an element but
(trace); a "-"
»
— —
tr
tr
0.39
• •»
0.30
tr
^ —
tr
tr
tr
tr
tr
were equivalent
denotes that no
1
Sn Ba
0.62 2.59
_ _
^ ^
^ ^
2.80
tr tr
^ ^
tr
^ ••
^ •
tr
^ ^
•. «.
^ ••
to less than the
X-ray counts were
** Minimum detectable limit =
*** Estimated 1.0 jjg/cm2 iron present
Background Counts
Peak Counts
X (Concentration)
-------
MEMORANDUM September 20, 1974
FROM: Charles T. Hare
TO: Vince Krause
RE: Project 11-3718. Contract 68-02-1230
Request for time extension at no additional cost
We would like to request a 3-month time extension at no additional
cost to the sponsor on Contract No. 68-02-1230. The reasons for this
request are:
1. technical difficulty with development of and contracting
for chemical analytical procedures;
2. longer-than-anticipated tests to acquire adequate
sample for analysis; and
3. late delivery of Government-furnished property
(ERG sampler)
The requested extension would move the end of the technical effort
to about January 22, 1975, and the end of the contract to about May 22,
1975. It is anticipated that this schedule can>be met if no further problems
develop.
cc: Ron Bradow
199
-------
Appendix B3.12
SOUTHWEST RESEARCH INSTITUTE
85OO CULEBRA ROAD • POST OFFICE DRAWER 28510 • SAN ANTONIO TEXAS 78284
September 9, 1974
TO:
FROM:
SUBJECT:
PREPARED
FOR:
Dr. Ron Bradow, Project Officer
Environmental Protection Agency
Research Triangle Park, North Carolina 27711
Charles M. Urban, Harry E. Dietzmann
and Karl J. Springer
Department of Emissions Research
Southwest Research Institute
San Antonio, Texas 78284
Monthly Progress Report No. 7 for the period August 1,
1974 through August 31, 1974; Contract No. 68-02-1275,
"Protocol to Characterize Gaseous Emissions as a
Function of Fuel and Additive Composition;" SwRI
Project No. 11-3902-001.
Environmental Protection Agency
Research Triangle Park, North Carolina 27711
Introduction
The purpose of this project is to develop test methods and over-
all test protocols to characterize potentially harmful gaseous emissions
from automobiles as a function of fuel and fuel additive composition. The
scope of this work includes the selection and application of analytical
techniques to automobile exhaust to determine various gaseous sulfur,
nitrogen and hydrocarbon compounds. These methods will then be applied
to the exhaust from two 350 CID Chevrolet engines, one standard and one
with a prototype catalytic converter, at the accomplishment of 1000 and
2000 mile intervals of operation under the LA-4 cycle conditions.
Progress During the Period August 1 through August 31, 1974
Phase I - Chemical Analytical Procedure Development and Demonstration
Sulfur Dioxide (SO2) - During the past reporting period, a number
of experiments were conducted to determine the extent of individual exhaust
gas component interferences. The results of these experiments are pre-
SANANTONIO H O U ST O N . C O R PU S C H R I S T I . T E X A S A N O W A S H I N G T O N O C
200 ,
-------
sented as Appendix A to this report. Two SO2 span gases (balance Air)
are in hand and once the manufacturer certified analysis is verified, a
sulfur balance will be conducted. Current plans call for continuous SO?
sampling during the 1975 LD FTP, with an on line integrator to deter-
mine average SO2 concentrations in each of the three portions of th<-
LA-4 test. Sulfur dioxide emission rates will be calculated in grams/
kilometer just as NOX, HC and CO.
It is planned to use air balance span gases and air zero gas as
the routine calibration gases for this instrument, since exhaust samples
will be air rich and relatively high in oxygen content. As documented
in Appendix A, using the Model 40 pulsed fluorescent analyzer in systems
with high oxygen concentrations (15-20 percent) minimizes any inter-
ferences that may otherwise create problems. The Model 40 SO2 instru-
ment thus far has performed satisfactory and the instrument may be
considered available for routine testing.
Sulfur Trioxide (SOg) - A satisfactory barium chloranilate column
has been prepared and the high pressure liquid chromatograph (HPLC)
system is deemed ready for routine testing. Exhaust samples have been
obtained and analyzed using the procedure developed by EPA Research
Triangle Park. A Beckman model 25 UV-VIS recording spectrophoto-
meter, equipped with a flow-through cell, is being used as the detector.
Several samples will be obtained during the qualification runs and sulfur
balance tests. Calculations for SO4~ emission rates will be made and
reported in g/km.
Although this procedure has been readily adopted for un-leaded
fuel, it is expected that an additional ion-exchange column will be neces-
sary to remove any anions in the extracted sample. Several references
to the original procedure developed at EPA have shown that anions such
as F , Cl , Br interfere with the barium chloranilate - SO^" reaction.
Non-Reactive Hydrocarbons (NRHC) - The NRHC system is ready
for routine testing and will be used during the qualification runs.
Polynuclear Organic Matter (POM) - During the last reporting
period, two fluoropore filters were loaded during four LA-4 runs. The
filters were extracted using a soxhlet extraction and reduced to a constant
volume and analyzed using the CAPE-7-68 DOAS instrument for aromatic
content. Using isokinetic sampling flow rates, a total of 125 ft was sam-
pled with the two filters. Assuming about a 6:1 dilution ratio, only 21 ft
of raw exhaust was actually obtained. In reviewing work performed using
the CAPE-6-68 polynuclear aromatic analysis of exhaust, it was apparent
that A tremendous quantity of exhaust was required for sample collection.
Two references Ui2)* which used the analytical methods developed under
the CAPE-6-68 program, used the entire exhaust rather than a small iso-
kinetic sample. These samples were collected using 3 blocks of 12 seven-
*Superscript numbers refer to list of References at the end of this report
201
-------
mode Lests, giving a total of 36 seven-mode tests per sample collected.
Also, these samples were collected using the entire exhaust stream
except for a small stream diverted for instrumental analysis. Using
this sampling procedure and testing sequence, it is estimated that well
over 4000 ft-' of raw exhaust was collected. In reviewing the overall
project requirements, it is estimated that by modification to the existing
sampling interface to include four filter systems and increasing the
sample volumes ten-fold, a total of over 40 I.A-4 runs would be required
for each test condition.
With the concurrence of the project officer, plans have been made
to delete the analysis of polynuclear organic matter (POM) from the
project performance plan. Retaining the analysis of POM in the per-
formance would severely delay the program schedule.
COS, and Lower Mercaptans - A short column of Chromo-
sorb 102 appears to be quite satisfactory for SO£ and COS using the
Flame Photometric Detector. Although dilute concentrations of COS
are readily made, considerable difficulty has been encountered with the
more polar and reactive I-^S. An all teflon system using Teflon sample
bags is best. Figure 1 shows a chromatogram of a 10 ml sample con-
taining 100 ppb each of HzS and COS.
Nitrogen Compounds - Efforts to obtain a column to do ammonia,
nitromcithane, and dimethylnitrosamine have been unsuccessful. The
Chromosorb 104 which looked hopeful for this purpose had an extremely
high bleed level and was impossible to use. Contact was made with the
manufacturer. They suggested some solvent washings which were tried,
but produced little improvement. The manufacturer is going to send a
new Chromosorb 104 from a different batch. It appears, however, that
chromatographic conditions for ammonia and nitromethane are not com-
patible. It appears the best approach is to use the Chromosorb 101 for
nitromethane and dimethylnitrosamine. Figure 2 shows a chromatogram
of 1 ng each of nitromethane and dimethylnitrosamine using the Hall
Electrolytic Conductivity Detector and Chromosorb 101 programmed
from 100°C to 170°C at 10° per minute.
Phase II - Setup, Conditioning, and Qualification of Fuel/Additive Test
Engines
During the past reporting, qualification testing has been completed
on test stand 2, the 1972 350 V-8 Chevrolet engine. A total of two cold
starts and four force cooled starts have been run and the results along
with the original contract specifications are presented in Table 1 and arc
summarized as follows:
202
-------
EPA
1975 LD FTP Specifications SwRI Emission Results
Emission Rates Max Mm Cold Start Force Cooled
HC, g/km 2. 5 1.5 2. 1 1. 9
CO, g/km 22 12 26. !J 22.2
NOX, g/km 2.5 1.9 1.7 1.7
These tests were conducted with the program test fuel doped with thio-
phene to obtain the target sulfur level of 0. 1 percent. As observed, the
range of HC (g/km) specified by EPA was 2. 5-1. 5 g/km and the cold
start and force cooled tests were both within the EPA emissions speci-
fications. EPA specified CO emission rates ranged between 22 and
12 g/km. The CO and HC emissions were only slightly higher in the
cold starts than in the force cooled starts. Due to the nature of this
project and the method of mileage accumulation, it is recommeded
that either force cooled or hot starts be used during Phase III emission
testing. Although there was no difference in the NOX emission rates
using both cold starts and force cooled tests, NOX was slightly less
than the minimum contract specifications.
The force cooled starts were conducted using a given set of pre-
determined conditions to insure repeatable test conditions. After the
initial cold start had been completed, the engine was shut off and cooling
water was forced through the engine. A fan was used to force lab air
over the engine and once the engine had been cooled down and remained
stable at not more than 3°F above ambient temperature, the engine was
ready for 1975 JL.D FTP testing. Similar force cooling procedures have
been used quite satisfactory in other projects.
An alternative procedure for conducting the Phase III emission
testing would be using a hot start rather than force cooled or cold starts.
A procedure could be readily devised for obtaining repeatable hot start
emission data should this approach be selected. One advantage of using
hot starts would be to enable the tunnel to be pre-heated to a consistent
temperature. This could aid in the preservation of sample integrity,
especially for the reactive sulfur containing compounds during the start-
up and initial warm-up period of the tunnel prior to the CVS. It is un-
certain as to the fate of SOz and SO3 if the tunnel wall temperature changes
significantly during the first 505 seconds of the 1975 LD FTP. This un-
certainty could be eliminated if hot starts are used exclusively during the
Phase III testing.
During the last reporting period, the GM-catalyst has been received
and installed on the prototype 1975 350 V-8 Chevrolet engine on stand 1.
At the present time, the catalysts are undergoing preliminary break-in
prior to qualification emission testing. It is expected that qualification
testing will be completed during the second -week in September. Once the
qualification testing is complete, a sulfur balance will be conducted on
203
-------
both engines using the 1975 LD FTP. During the sulfur balance tests
sulfur dioxide and sulfur trioxide (sulfate) emission levels will be
obtained. It is not expected that I^S, COS or mercaptans will signi-
ficantly affect the sulfur balance. Once the qualification testing and
the sulfur balance has been completed, both engines will be ready for
mileage accumulation.
Current Problems
Insufficient sample acquisition of polynuclear organic matter
on a filter media during a reasonable number of LA-4 tests has deleted
this analysis from the testing schedule. Analysis of HzS, COS and
nitrogen compounds is progressing slowly and hopefully will be ready
for routine analysis once the qualification runs have been completed.
Work to be Performed During the Next Reporting Period
Qualification testing on both engines is expected to be complete
during the next reporting period. A sulfur balance will be obtained on
both engines during LA-4 testing to validate both the SO2 and SO, sam-
pling systems. It is expected that full scale demonstration of all sampling
and analytical methods will be accomplished, as well as initiation of
mileage accumulation on the first additive package.
Prepared by:
Submitted by:
Charles M. Urban
Senior Research Engineer
Department of Emissions Research
Karl jy Springer
Director
Department of Emissions Research
and
l~
Harry E. Dietzmann
Senior Research Chemist
Department of Emissions Research
204
-------
'.:• ^FIGURE-r.
SEPARATION OF H2S AND COS
-------
-------
TABLE 1. PRELIMINARY QUALIFICATION EMISSION RESULTS
(1972 350 CID CHEVROLET ENGINE - STAND 2)
1975 LD FTP
Emission Rate, g/km
Type of Start Date Run HC CO NOX
Cold Start 8-27-74 1 2.06 26.16 1.74
Cold Start 8-28-74 1 2.05 26.43 1.60
Force Cooled 8-27-74 2 1.82 21.99 1.58
Force Cooled 8-27-74 3 1.79 23.54 1.71
Force Cooled 8-27-74 4 2.00 21.84 1.84
Force Cooled 8-28-74 2 2.05 21.27 1.62
207
-------
LIST OF REFERENCES
1. Gross, George P., "The Effect of Fuel and Vehicle Variables
on Polynuclear Aromatic Hydrocarbon and Phenol Emissions,"
SAE paper 720210, presented at the SAE Automotive Engineering
Congress - Detroit, Michigan - January 10-14, 1972.
2. Hoffman, C. S., et al, "Polynuclear Aromatic Hydrocarbon
Emissions from Vehicles, " presented before the Division of
Petroleum Chemistry, Inc., American Chemical Society, Los
Angeles Meeting, March 28-April 2, 1971.
208
-------
APPENDIX A
MODEL 40 PULSED FLUORESCENT
SO2 ANALYZER INTERFERENCES
209
-------
A-2
MODEL 40 PULSED FLUORESCENT SO2 ANALYZER INTERFERENCES
The use of the model 40 pulsed fluorescent SO2 analyzer in
the presence of other exhaust gas components could cause potential
interference problems. The intended use of this instrument will be
in sampling CVS exhaust on a continuous basis. Since the exhaust
will be essentially an air-based sample, the experiments described
are orientated toward this particular application.
CO and CO2 Interferences
Initial CO and CO2 interference checks were conducted using
a single bottle cart containing eight golden standards named by EPA
Ann Arbor. This group of bottles contained multi-component blends
as well as single component mixtures. All eight bottles contained
nitrogen as a balance gas. The first set of data was obtained using
standard regulators with neophrene diaphragms and teflon tubing. The
results of these tests are found in Table A-l. Response as SO2 ranged
TABLE A-l. MULTICOMPONENT BLENDS OF CO AND CO2/N2
RESPONSE AS SO2 IN MODEL 40 PULSED FLUORESCENT
ANALYZER (NORMAL NEOPHRENE DIAPHRAGM REGULATOR)
Concentration, %* Response as
Test CO CO2 ppm SO2
1 9.58 5.70 15.5
2 - 14.16 10.3
3 - 12.36 16.0
4 - 11.22 16.5
5 5.39 10.46 17.5
6 2.79 13.18 17.0
7 1.39 - 16.5
8 0.48 15.23 17.5
*balance gas N2
from 10. 3 to 17. 5 ppm with no apparent correlation for response as ppm
SO2 and interference concentration. The CO concentrations varied from
0.48 to 9.58 percent, while the CO2 values ranged from 5.7 to 15.23 per-
cent.
Since it was obvious that something other than a straightforward
single compound interference was involved, additional experiments to de-
termine the extent of other variables that might lead to apparent inter-
ferences were conducted. The first such variable checked was the effect
210
-------
of regulator diaphragm type on the response as ppm SC^. Since other
data was previously run and substantial data available for neophrene
regulator diaphragms, it was decided to investigate the response of
several of the previously tested bottles using metal diaphragm regulators
like those used for hydrocarbon span gases. The results of this experi-
ment are found in Table A-2. Although it is difficult to make any defi-
nite conclusions, it was observed that the response as ppm SC>2 for these
same three CO/CO2 bottles was 2. 5 to 3.0 ppm less for the metal dia-
phragm.
TABLE A-2. MULTICOMPONENT BLENDS OF CO AND CO2/N2
RESPONSE AS SO2 IN MODEL 40 PULSED FLUORESCENT
ANALYZER (METAL DIAPHRAGM REGULATOR)
Concentration, %* Response as
Test CO CO2 ppm SO2
1 9.58 5.70 13.0
2 5.39 10.46 14.5
3 0.48 15.23 15.0
^balance gas N2
The next item checked was CO2 in balance zero air. A clean
Tedlar bag was prepared with a double end shut off quick connect and
filled with zero air. The bag sample was then analyzed in the Model
40 SO2 instrument and no response was observed. The bag was then
doped with some pure CO2 to give a CO2 concentration of about 13 per-
cent. The bag was then run in the SO2 instrument and still no response
was observed. As a result of this experiment, it was obvious that CO2
alone could not be considered to be an interference compound; however,
in conjunction with other species could present interference problems.
At this point, two facts were apparent; first, CO2/N blends gave 10-16
ppm SO2 response and secondly, CO2/Air blends gave no response.
To determine the extent of the CO2/N2 interference, additional
experiments were conducted. The availability of a range of O2/N2
blends was used to narrow down this problem. The bottles were N2
zero gas, 5 percent O2/95 percent N2, 10 percent O2/90 percent N2,
15 percent O2/85 percent N2, and 20 percent O2/N2- Several fresh
bags were prepared and each blend was analyzed for response as ppm
SO2. Then each bag was doped with pure CO2 to a level of about 10 per-
cent. These bags were then run and the results of these tests are found
in Table A-3.
It was apparent that by running the oxygen-nitrogen blends with-
out any CO2, certain effects could be observed. As the amount of oxygen
in the sample decreased (and the nitrogen concentration increased), a
211
-------
A-4
positive response as ppm SC>2 was observed, even though no other compounds
were known to be present. When the blends of about 12 percent CO^
in various O2/N2 ratios were analyzed, it was found that the CO2 and
03 acted much the same in that the sum of the CO2 and O2 concentra-
tions had the same quenching effect as the O2 concentrations alone.
TABLE A-3. MUJLTICOMPONENT BLENDS OF COz/Oz/^Z
RESPONSE AS SO2 IN MODEL 40 PULSED FLUORESCENT ANALYZER
Concentration, %* Response as
Test CO2 O2 N2 ppm SO?
1 - 0 100 10+
2 5 95 0.5
3 10 90 0.2
4 15 85 0.1
5 20 80 0.0
10+
0.5
0.2
0.1
0.0
2.0
Initial conclusions regarding CO and CO2 interferences indicate
that these two exhaust species do not interfere as positive SO2 response
provided there is a sufficient quenching effect provided by oxygen in the
sample. Problems could be present if direct exhaust samples are ob-
tained and oxygen levels are low. Preliminary experiments indicate
that oxygen levels above 5 percent have less than 0. 5 ppm response as
SO2. In cases where a CVS air diluted sample is obtained, no inter-
ferences due to CO or CO2 were observed.
NOX Interferences
Five bottles of NOX/N2 were used to conduct initial NOX interfer-
ence experiments. These were also golden standards named by EPA
Ann Arbor. Although these bottles were named as NOX, they were actually
NO in N2 cylinders as verified by chemiluminescent analysis. The con-
centrations were selected to be typical ranges that might be expected in
1975 FTP testing. The results of this test are found in Table A-4. NOX
concentrations ranged from 42 to 220 ppm and the apparent interferences
as ppm SO2 varied from 17.0-36.0. This was the only gas tested which
appeared to produce an increased response with increasing component
concentration. It should be noted that N2 zero gas produced some 10+
ppm response as SO2.
212
6
7
8
9
10
11
10%
10%
10%
10%
10%
100%
0
4.5
9.0
13.5
18
-
90
85.5
80.0
76.5
72
-
-------
TABLE A-4. MULTICOMPONENT BLENDS OF NOX/O2/N2
RESPONSE AS SOz IN MODEL 40 PULSED FLUORESCENT ANALYZER
1
2
3
4
5
6
7
8
9
10
11
12
Concentration
Test NOX, ppm
42
78
95.5
133.5
220
140
400
550
710
1125
1400
1750
02.
21
21
20
22
20
19
20
N2>
100
100
100
100
100
79
79
80
78
80
81
80
Response aR
ppm SO 2
17.0
24.3
27.0
32.0
36.0
0. 1
0
0.25
0.
negative
negative
negative
There was a definite trend observed regarding NOX concentration
as a function of response as ppm SO2, as shown in Figure A-l. Previous
experiments involving CO and COz interference checks indicated that
the presence of nitrogen and the lack of oxygen could lead to apparent
interferences. With this in mind, several blends of NO/N2 were diluted
with oxygen to obtain a nominal 20% O2« The conversion of NO to NO2
was immediately apparent due to the color change of the NO-*-NO2 reaction.
Although the previous NOX check involved NO/N2 blends, this experiment
actually was NO2/Air and comparison is somewhat difficult. The con-
centration of NOX ranged from 140-1400 in the bag samples analyzed. The
O2 and N2 concentrations were relatively the same for purposes of this
experiment. At any rate, the low concentrations (140-700) of NO2/Air
produced only slight response as ppm SO2- At higher concentrations of
NO2, a negative response was observed for several gases.
It is difficult to make any absolute conclusions based on the data
presented in Table A-4. Although NO/N2 blends do give a positive response
as ppm SO2, it is impossible to determine the extent of NO/Air interferences
due to the NO -*NO2 oxidation in air. Bag samples obtained from a CVS
are significantly air rich and have 03 concentrations above 15 percent under
most conditions. Since the CVS bag samples contain relatively low concen-
trations of NOX diluted in air, it is not felt that any significant NOX inter-
ferences will be expei ieuced.
213
-------
A-6
rg
8
a
a
u>
U>
o
a
(0
2 + N£ only
°2 + N2 Plus
120 ppmC benzene
140 ppmC toluene
) 5 10 15
Percent Oxygen - Balance Nitrogen
Figure A-l. The Effect of Benzene and Toluene in
Various C>2/N2 Blends as pprr. SC>2 in
Model 40 Pulsed Fluaresc.t-nt Analyzer
214
20
-------
HC Interferences to Model 40 Pulsed Fluorescent SOz Analyzer
Several experiments were conducted using typical hydroc . rbon
blends in N£ and air. The initial tests were conducted using p< opar.e
in N2 and propane in air. The results of these tests are found in Pa,? e
A-5. The golden standard span gases were originally thought to be air
TABLE A-5. PROPANE SPAN GAS RESPONSE AS SO2
IN TECO MODEL 40 PULSED FLUORESCENT ANALYZER
Concentration, Balance Response as
Test ppmC Gas ppm SO 3
_ __ Air o^Q
2 34 NZ 14.0
3 168 Air 0.0
4 301 Air 0.0
5 1024 Air 0.0
based gases, but during the tests it was found that the 34 ppm C bottle
was actually a balance N2 gas. This accounted for the fact that all of
the other propane in air gases gave no response, whereas the 34 ppmC
/N2 8as 8ave an apparent response of 14 ppm SO2. Hydrocarbon con-
centrations, varying from 25 to 1024 ppm C balance air, were found to
produce no response as ppm SO2.
It is suspected that the balance N2 was responsible for the apparent
interference in the 34 ppm C bottle. Once it was verified that typical air
based HC span gases produced no interferences, it was decided to check
the Model 40 pulsed fluorescent SO2 instrument response to aromatic hydro-
carbons. Two aromatic hydrocarbons typically found in automotive exhaust
were selected for this experiment. These were benzene and toluene. The
availability of several gases containing various ratios of O2 and N2 were
selected for these tests. Baseline readings were obtained on each of these
gases and these results are presented in Table A-6. A bag sample of
each of these gases was obtained and a predetermined amount of benzene
and toluene were added to each bag. Nominal benzene concentration was
120 ppm C and toluene concentration was about 140 ppm C.
The results of these experiments are illustrated in Figure A-2. In
comparing the response as ppm SO2 to the base O2/N^ blends to those same
blends with added benzene and toluene it is apparent that some sort of inter-
ference due to aromatic compounds is present. It almost appears that the
inte* . ence found in this test is an exponential function. Initial conclusions
from this interference check indicate that samples containing less than 5
percent O2 can have significant interference. These evaluations were con-
ducted on the 0-10 pp» -<~ale and the maximum interference that might be
expected during CVS «. ration would be 0.2 ppm or 2 percent of full scale.
215
-------
A-8
40
35
30
CO
O
CO
a
a,
n
25
20
to
a
o
CL.
CO
a>
15
10
str
ztnr
1ZO
T40
NOX Concentration, ppni
Figure A-2. The Effect of NOX Concentration (balance
on Response as ppm SO2
216
-------
A-9
TABLE A-6. MULTICOMPONENT BLENDS OF BENZENE,
TOLUENE/O2, NZ RESPONSE AS SOz
IN MODEL 40 PULSED FLUORESCENT ANALYZER
Concentration, ppm C Concentration. % Resp • .e as
Test Benzene Toluene OZ NZ ppm SOz
1 - 0 100 10+
2 - 5 95 0.5
3 - 10 90 0.2
4 - 15 85 0.1
5 - 20 80 0.0
6 140 160 0 100 10+
7 140 160 5 95 0.8
8 140 160 10 90 0.4
9 140 160 15 85 0.2
10 140 160 20 80 0.15
General Comments of Model 40 Pulsed Fluorescent SO? Analyzer
Upon completion of the aforementioned experiments, several
contacts were made with other individuals who had working experience
with the instrument or was involved with Thermo Electron Corporation.
The first contact was Glenn Reschke at General Motors. He had conducted
numerous experiments with this model instrument, many of the same
nature of the SwRI evaluations. Although his particular application was
for use in direct automotive exhaust sampling, his conclusions regarding
the various component interferences were essentially identical to those
presented herein.
Further verification of individual component interferences conclu-
sions was obtained from Dennis Helms of Teco. He re-iterated the items
presented in this report and those indicated by Glenn Reschke. Recom-
mendations for specific application to CVS type exhaust sampling have been
previously incorporated into the exhaust sampling system.
It may be considered a concensus of opinion that sampling from any
air-rich CVS system and using air balance SOz span gases and air zero
gases minimize on potential interferences. Should direct exhaust sampling
with relatively low oxygen concentrations (less than 5 percent), additional
interference checks might be warranted.
217
-------
Appendix B3.13
Status Report
ROAP 26 AAE
Task 008
Validate Engine Dynamometer Test Protocol
for Control System Performance
Concept:
The initiaj concept underlying this objective was to test a control
systems protocol being developed under contract at the Dow Chemical
Company. These tests would confirm the ability of the test to measure
the influence of fuel additives on various kinds of automobile emissions
control devices including oxidation catalysts, EGR NO control devices
A
and others. Resources available for the task (0.5 man-year and $15,000)
were judged too small for complete development of a protocol.
In fact the Dow data arrived too late (July 1974) to have any impact on
this portion of the in-house program, and experimentation was begun on the
basis of very preliminary results from both Bureau of Mines and Dow. The
Dow report has concluded that the candidate protocol was not satisfactory
for assessing catalyst degradation, and this fact has had substantial
impact on the in-house program in the last three months. Subsequent dis-
cussion will outline the problems associated with Control System degradation
measurement, the candidate methods, and analysis of available contractor
data on regulated emissions. Attachments include a report of the engine
dynamometer hardware development problem including information on the subject
of speed-load regulation, the analysis of contractor data, and a draft test
protocol.
Protocol Development:
Early in the Dow contract a number of facts became apparant. First,
individual components of an overall control system probably should not be
tested individually. Catalysts, EGR packages, evaporative loss canisters, etc.,
218
-------
are engineered to constitute an integrated pacakge system. Operation
of the components is so interrelated that it is probably impossible to
isolate the additive effect on EGR etc., separate from its overall effect
on the system. Furthermore, there is essentially no information on the
level of detrimental effect on such peripherals as PCV values, ECR, or
evaporative canisters. Therefore, the required precision of testing
required to isolate such effects cannot be assessed a priori1.
Such is not the case with catalysts. Catalyst degradation by lead,
phosphorous, and sulfur compounds originating from fuel or lubricant
additive packages has been much studied. The level of effects from known
catalyst poisons and the mileage accumulation required to discover them
is fairly well established. Thus, low levels of tetraethyl lead can cause
significant decreses in catalyst activity in a few thousand miles. Sulfur
and phosphorous effects require 20-30 thousand miles before degradation
2
becomes important.
Since the influence of fuel additives on control peripherals is ex-
pected to be low and difficult to measure, any protocol to establish these
minor effects must be either capable of high precision or rely on very
extensive replication. Since these peripherals are part of an integrated
system, the only rational choice is a test of the whole car with the
variability that entails. Attempts to develop a high precision engine
dynamometer test based on the present Urban Driving Cycle in the contract
program were not successful in reducing the variability inherent in the
basic system. Therefore, at present, it appears that only a very large
automobile fleet test can detect these effects. At present the cost to
perform such testing cannot be justified in terms of positive impact on ambient
air quality for presently regulated emissions.
Therefore, concentrated effort was placed on determination of catalyst
deterioration. The literature indicated that unregulated consumer driving
can produce recognizable catalyst deterioration. However, the variability
in FTP testing makes detection of effects very .difficult. Thus, on the
basis of statistical analysis of Dow, Bureau of Mines and Calspan data, it
was computed that the minimum number of car trials needed to reliably establish
a 10% loss in catalyst activity is 86 in FTP testing.
219
-------
Detection of a 25% loss in activity as a pass/fail criterion would require
a minimum of 16 replications each of additive and control cars.
Mileage accumulation is also a serious problem. Catalyst deterioration
occurs more rapidly at higher mileage values. On the basis of literature
studies it appears that at least 25,000 miles is required to establish a
25% deterioration of catalyst activity for either CO or hydrocarbon.2'3
Description of the Protocol;
Catalyst activity can be determined most precisely by using a conven-
tional engine operated at steady state, 30 mph level road load as a source of
standard exhaust. Exhaust hydrocarbon level must be carefully controlled by
continual monitoring of exhaust composition and carburetor adjustment. Thus,
it is proposed to initially rate a series of catalysts on the dynamometer
stand and instill these catalysts in test cars for mileage accumulation of
25,000 miles. At the end of the mileage accumulation phase the catalysts
would be removed and re-rated on the dynamometer stand.
Catalyst performances rating parameters are % efficiency in hydro-
carbon and CO removal, time from cold start to minimum hydrocarbon level, and
rate of catalyst bed temperature rise. The test sequence suggested is
similar to that used by auto manufacturers in screening studies and is
generally referred to as the AC 813 test.
At present there is insufficient information from our in-house testing
to establish the repeatibility of the test. If FTP testing is done in addition
to the AC-813 test, the ability to detect a 25% increase in HC or CO can
be controlled.
Conclusion:
A control system performance protocol is recommended which can control
the chance of detecting a detrimental effect due to the additive at 25%
using FTP testing. A second more sensitive test is proposed which is
potentially capable of detecting much lower deterioration factors. Since
this test is not as yet standardized, data from this source can only be
used to indicate the possible pressures of small effects.
220
-------
Appendix B3.14
FUEL ADDITIVE PROTOCOL DEVELOPMENT
Analysis of Bureau of Mines Dal.*
Control Device and Gaseous Emissions Protocol Development Program
by Ronald L. Bradow, Chief, ETCS
I. Introduction
The gas emissions protocol development program at Bureau of Mines
was designed to test the efficiency of a test schedule in assessing the
influence of two fuel additives (Chevron F-310 and Ethyl AK-33-X) C"1
emissions. A second function involved development of methods for de-
tecting and determining any new toxic products resulting from the use Qf
those additives. A report on this project from the bureau covers well
the largely negative results of the search for new compounds, ilcvcver,
there is little discussion of the efficiency of the test schedule and
procedures in assessing additive effects. This paper discusses the
available gaseous emissions data on a statistical basis and points out
the many shortcomings of the test program. Alternate approaches to the
problem of assessment of additive effects are also discussed.
II. Test Schedules
The program seeks to evaluate fuel additive effects for non-catalyst
cars. It was assumed that an engine dynamometer test stand could achieve
higher reliability and repeatability in assessing additive effects than
could be done with a fleet of cars. Therefore, the program required
development of an engine dynamometer test procedure of high reliabiljty.
The procedure initially specified an elaborate break-in sequence involv-
ing a series of steady state operations designed to seat rings and wear
in bearings, followed by a conditioning sequence of LA-4 routes to stabi-
lize deposits. Engine dynamometer operation involved mileage accumulation
by either repeated urban driving cycles or by the durability driving
schedule with intermittant testing by the 1975 Federal test procedure.
The cars were driven on the road in ordinary consumer driving with no
specified cycle. At 1,000 mile intervals 1975 Federal test procedures
were run on the cars.
The goal of increased test reproducibility with engine dynamometer
tests was not achieved by any of the groups operating this procedure. In
fact mi-smatch of vcnicles both initially and during mileage accunulation
was so poor as t;i n-ahe cross-car comparisons of littJe value. In the-
in-house cxpcrirnerLs it has been found that careful raw exhaust measure:r,u::it
and attention to c,;Lburetor fuel flow adjustments at idle and at several
221
-------
steady state speeds is necessary to assure adequate vehicle and engine
matching.
III. Analysis of Bureau of Minos Gaseous Emissions Data
Table I presents values from 1975 FTP emissions tests at the
beginning of the test program, just before beginning additive injection,
and the mean of all base fuel tests. It is immediately apparent that
both stationary engines are lower in all emissions than the cars. The
AK-33-X car appears somewhat higher in hydrocarbon and CO than the base
car and this point will be developed statistically later. Both methane
and ethylene emissions are also low in the stationary engines relative
to the base car and the methane/ethylene ratio is also lower on that
basis. Detailed consideration of the g.c. analysis data suggests a
considerable effect of fuel composition on diese results. Initial
runs were performed with indolene fuel with a 35 vol.% aromatics concent.
Later the EPA reference fuel with a 24% aromatics content became avail-
able and a switch to that fuel was made. The switch v/as attended with
a 35-40% decrease in methane, benzene and toluene in the base and F-310
car and a 15-20% decrease in these components in the apparently richer
AK-33-X car. During mileage accumulation with the additives there was
no apparent increase in any specific hydrocarbon emission. Variability
in aldehyde data v/as high and no apparent differences were detectable
on inspection.
Using the data in Table 1 and test variances for the control car,
it is possible to decide statistically if the cars were originally
matched. To test this point values of the Z statistic (distributed as t)
were computed. Table II presents these values for comparisons of the test
cars and engines with the control car. It is clear that the F-310 car
and control car were matched, but the AK-33-X car has significantly
higher CO and hydrocarbon emissions and must, therefore, be operating
somewhat richer than the other two. The engine emissions were in every
case lower than those of the cars and clearly, comparisons between
engines and cars are invalid. It appears that the inertial loadings
were too low for the engines and comparisons of peak horsepower or torque
under some standard acceleration would be helpful in establishing this
point.
The variance in basic emissions data on both engines and vehicles
was similar to that previously reported by ECTD-MSPCP. Table III gives
values of means, standard deviations and comparisons of means for all
additive runs with test cars, and for AK-33-X runs with stationary engines.
Again the car data is every where greater than that from the stationary
engines. Standard deviations of the present tests are shown compared with
222
-------
those previously reported for fourteen similar cars by EPA-Ann Arbor.
Generally variance of both car and engine tests are similar to the
literature data.
Part B of Table III presents comparison of mean values for car
tests using the t statistic. The most important conclusion to L
derived from this data is that the AK-33-X car has significantly
greater hydrocarbon emissions than either the F-310 or base cars at
the test end. Of course this was also true at the beginning of uhc L -FT
cycle. If the initial and final values for that car are compared, it
was found that there is no significant difference in hydrocarbon but
the carbon monoxide is significantly lower at the end of the AK-33-X
test. Thus, it is impossible to assign any clear effect to either
additive on the basis of these car tests. Since only two individual
engine tests were conducted to high mileage, it cannot be definitely
established that, during mileage accumulation, by sheer chance both
engines changed mechanically to produce the observed test result.
Therefore, oven though the difference between the base fuel and
additive fuel tests have statistical significance, the significance
cannot ascribed uniquely to the additive. Clearly, either sufficient
control data must be obtained to assure that the changes observed
cannot be ascribed to mechanical changes or sufficient mileage experiments
must be run to allow removal of vehicle or engine influences.
It is important to note that none of the Bureau of Mines, EPA or
Dow engine dynamometer testing has found any improvement in repeat-
ability over vehicle testing. Comparison of the variances of test data
can be made in this case using the F statistic and Table IV prescnrs
the data. In no case v/here any of the Bureau of Mines tests significant-
ly different in variability than literature 1975 FTP vehicle tests.
Table V presents a comparison of all stationary engine data indicat-
ing the similarity of F-310 and base fuel data on both engines.
Table VI presents comparison data from Calspan, Dow and EPA-RTP.
As in the Bureau of Mines tests, the individual automobiles have a wide
variety of emissions characteristics. The Calspan cars are significantly
mismatched as were the initial tests with the EPA cars. Since tnese £1"A
tests were run, a considerable effort has been expended in maintaining
the repeatability of the vehicles. This has involved periodic determina-
tion of raw exhaust composition and, hence, fuel air ratio at idle, 15, 30,
45, and 60MPII steady state conditions.
223
-------
Since the influence of fuel additives on emissions may be in-
direct; for instance, the additive may increase or decrease fuel-air
ratio, the additive effect is indistinguishable from mechanical changes.
A test with few vehicles may be very difficult to control with the ir.ost
careful maintenance. Considering th" test variability, it is possible
to calculate the minimum number of tests required. to_discover a real
difference of any given percentage in hydrocarbon'NOX emissions. To
find a 10% increase because of the additive, a minimum of 86 tests is
required for both the additive and control cars. For a 5% increase
344 tests are required.
Conclusion
The variability in all gas analysis data in the fuel additive contract
program strongly suggests that an engine dynamometer FTP protocol cannot
adequately assess the minor influences which fuel additives are likely
to exert. Only gross effects (in the order of 30% or greater) in the
emissions values can be detected, by the certification test procedure in
a reasonable number of replications, say 10 or fewer.
In order to successfully test the additive effect, the test itself
must be greatly simplified possibly to the point of simply determining
steady state fuel-air ratios. The only alternative would seem to be a
large fleet test of 100 cars or more.
224
-------
TABLE I
Analysis of Bureau of Mines Initial Emissions Data
Final Report EPA-IAG-097(CD)
grams/mile
Vehicle/Fuel
Hydrocarbons
Initial
Begin additive
Mean Value a
N
CO _
Initial
Begin additive
Mean value a
N
NOx (corr.)
Initial
Begin additive
Mean value a
N
Methane, ppmc
Initial
Begin additive
Ethylene, ppmc
Initial
Begin additive
Aldehyde _
Initial
Begin additive
* 4550 miles
# 4950 miles
0 3990 miles
a mean of all preadditivc runs
Base
2.92
2.78*
2.72
5
46.7
63.6*
56.9
5
4.62
5.60*
5.23
5
16.86
15.04
17.79
17.02
F-310
2.76
2.77
2.72
5
59.5
62.1
64.5
5
4.55
5.81
5.16
5
17.39
12.81
19.95
18.81
AK-33X
3.09
2.92
3.14
6
74.4
63.5
71.6
6
4.85
5.16
5.49
6
17.49
14.61
19.75
23.32
Sta.
A
2.18
1.29#
1.38
6
22.0
17.5
20.8
7
3.42
2.86
3.21
7
7.23
7.85
11.46
10.99
(6,400
Engine
B
1.37
1.79°
1.59
4
18.2
20.5
17.8
4
1.79
2.86
2.46
4
7.11
9.70
9.70
mi.)
0.103*
0.086
0.088
0.074
0.109
225
-------
TABLE II
Comparison of Initial Car and Engine
Emissions Data
Basis - Control Car
Statistic O—
HC
CO
NOX
/ 1/N]
D.F. i
Value
0.32
9.9
0.71
L + I/Ha
f
of Z for
F-310
0
1.20
0.154
0.64
8
1.86
Z*
AK-33-X
2.18
2.47
0.610
0.60
9
1.83
A
7.22
6.29
4.83
0.58
9
1.83
B
5.27
5.89
5.82
0.67
7
1.89
significance, 95%
confidence level
-X
1/N2
D.F. = N! + N2 = 2
226 ,
-------
TABLE III
A. Variance of Bureau of Mines Data-additive Runs
Car
Bumines
EPA 1/2
Bumines
EPA !'2
Bumines
EPA l'2
or Engine
HC
X
cr
-------
TABLE IV
Comparison of Variance - BuMines and EPA Data
a.) Additive Runs
EPA data from Ref. 1 - N = 28, N = 10
F5% - 2.75
Values of the F Statistic
HC
CO
NO
Base
1.0
1.20
1.45
F-310
1.71
1.16
1.37
AK-33-X
1.07
1.41
1.72
Eng.A*
1.68
1.73
1.01
Eng.B *
2.34
1.22
1.34
* F-310 runs - corrected for level differences by assuming
constant relative o~
228
-------
TABLE V
Stationary Engine Data
Additive Fuels
Engine Base F-310 AK-33-X
A B A B A B
HC
X 1.63 1.78 1.56 1.69 2.10 2.27
er- 0.29 0.24 0.12 0.093 0.44 0.44
CO
X 22.1 26.9 25.7 38.8 19.0 29.0
-------
TABLE VI
1972 Chevrolet Engine and Vehicle Emissions Data
Base Fuel
A. Calspan Data - 3 cars
DOW
ABC Engine Dyno
HC
CO
NO
X
X
cr
N
X
-------
Appendix B3.15
PROPOSED EPA PROTOCOL
CONTROL SYSTEM PERFORMANCE
1. Scope.
The Administrator* United States Environmental Protection
Agency, if he determines that a fuel additive offered in commerce
has a significant probability of causing deterioration of control
System performance, may require of the manufacturer of such an
additive to perform the tests contained within this protocol.
Under the provisions of the protocol, a fleet of 16 matched
o
-------
cars and tested by the chassis dynamometer procedure once more.
Data from certification and dynamometer testing shall be sub-
mitted to ZPA for determination of detrimental effect of the
additive.
The provisions and dimensions of the test protocol v/ere
selected by statistical considerations. On the basis of know-
of
ledge/the variability inherent in Federal Certification Testing,
it is possible to control the probability of finding false
positive and false negative responses to the protocol. It has
been determined that the probability of errojs of both the first
and second kind shall be controlled to 5% if the catalyst
deterioration response is as great as 25%. The minimum number
of pairs of vehicles required for a statistically valid ex-
periment on this basis is 16.
2. Submission of Car Fleets.
The manufacturer shall select nev; passenger cars in pairs
to comprise the additive test fleet. The cars shall be selec-
ted to be fairly representative of the current year's pro-
duction catalyst-equipped cars and shall include V-8, in^line-6,
and 4 cylinder engines in a variety of vehicle weights. Cars
may be immediately tested; however, the proposed vehicle fleet
selection must be submitted to EPA for approval. The approved
car fleet must be then tested by the Federal Certification
Test Procedure for light-duty vehicles (Federal Register. Nov. 12,
1972 as ammended by Federal Register. Jan. 14, 1974) with the
following exceptions; Since the vehicles will have been already
certified, no Part I submission of characteristics need be
made; the evaporative omissions tests, heat-build-tost, and
preliminary vehicle preparation shall be omitted, except that
232
-------
a preparative Federal Urban Driving Cycle shall be driven on
the day immediately before the certification test. A series
of throe consecutive certification tests shall be performed
with each vehicle. The data from these initial tests shall
submitted to EPA for determination of the comparability of
additive and control cars. On receipt of approval of com-
parability, the additive manufacturer shall then remove th"
catalytic converters from each2ar and perform a second engine
dynamometer test for converter efficiency as described below.
3. Converter Efficiency Test
A V-8 engine of approximately 350 cubic inches displacement
manufactured by an American-based automobile manufacturer as a
production engine in the model years 1972 or more recently
shall be used as an exhaust generator and shall be mounted on
an engine dynamometer test stand. " The dynamometer may be of
the hydraulic, eddy-current, or electric motor-generator types.
The dynamometer shall be capable 'of absorbing, measuring, and
controlling the engine load to'* 1.0 ft-lb. of torque at simu-
lated 30 rnph level road load for that engine, mounted in a
vehicle of /fOOO Ib. intertia weight equivalent. The level
road load chosen shall not be less than 15.0 nor more than
20.0 ft-lbs of torque. The engine shall be tuned and operated
so as to producOjafter 15 minutes of warm-up operation,an ex-
haust gas corresponding to the follov/ing specifications:
CO - 0.4 to 0.6 Vol. %
Hydrocarbons(FID) - 500 to 750 ppm C vol./vol.
02 - 2 to 3 vol. %
Temperature - 730 + 30° F
Catalytic converters sh ill bo mounted in the exhaust system
of this engine in a ™annor consistent v/ith their use in a vehicle.
233
-------
Thus, catalysts designed to treat the whole exhaust of a 300 cu.in.
or greater engine shall be mounted so as to treat the whole test
engine exhaust. Catalysts meant for ^ or 6 cyclinder engines or
for one bank of a V-8 shall be mounted in the exhaust of one
bank of four cylinders of the test engine. A by-pass section
of exhaust pipe, controlled by means of coupled diverter valves
at either end, shall be constructed around the converter such
that the engine nay be warned iip initially and its exhaust
then diverted through the initially cool converter. Tosts of
hydrocarbon and CO content of the pre-catalyst exhaust shall be
made by drawing a sample of exhaust from the exhaust pipe sec-
tion immediately preceding and immediately following the converter.
Samples of gas may be obtained by use of a metal bellows pump,
an ice-bath water knock-out trap, stainless steel transfer
lines and fittings. Samples of dried exhaust gas shall be
pumped directly to approved CO, COp, and hydrocarbon analysis
instruments. Analysis for CO and.C02 shall be conducted using
non-disperive infrared analyzers and analysis for hydrocarbons
by flame ionization analyzers; all such CO, C02, and hydrocar-
bon analyzers shall conform to the specifications set forth in
the Federal Register standards for light-duty vehicle certi-
fication (Federal Sorristo*'. Nov. 12, 1972).
• The gas transfer system shall be capable of a flow rate
sufficiently great that a sample of raw exhaust shall have a
residence time in the combined transfer and analytical system of
no greater than 1 second. The overall analytical system shall
be capable of a 00;"j response to a 10 ppm pulse in hydrocarbon
concentration in the rav: exhaust in 3 seconds. Records of
tost results ehail be fln(:c for cach tost using strjp chart
234
-------
recorders or othor data-logging devices capable of information
up-ciate every 10 seconds or faster.
k. Dynamometer Test Procedure.
The engine sha^l be started and operated on the test iiie'1
at the test condition for 15 minutes or core to assure stabili-
zation of cx'::aust conditions. To begin the converter performance*
test, diverter valves are sv/itched, allowing exhaust to enter
the converter. The initial time, catalyst temperature, and
exhaust analysis before and after the converter are recorded.
Engine conditions are maintained at 30 mph, level road load,
for 600 seconds during which tine exhaust composition and
catalyst temperatures are recorded. At the conclusion of the
test the engine nay be shut down and the catalyst may be
cooled by drawing room air through the converter. Successive
experiments shall not be run before the catalyst temperature
drops below 130° F.
The results of 6 such tests for each converter shall be submitted to EPA in the
form of data tables of computed 'percentage breale-through of
hydrocarbons and CO and certified copies of strip chart or
data-logger records indicating time to minimum values.
5. Fuels.
Two types of fuels shall be used in the additive tests,
certification and mileage accumulation fuels. All gasoline
used in engine and vehicle testing shall conform to the speci-
fications given in the Fcdcra1 PC-Bister. Nov. 12, 1972, except
that the Reid vapor pressure shall bo between 9 and 10.5 psi.
The mileage accumulator* fuel shall conform to SAE recommended
practices for distillatio:, and Peid vapor pressure characteris-
tics in the clinatc arcr.-. in which the vehicle tests are operated.
235
-------
Other specifications for mileage accumulation fuels are given be-
low:
Table 1
Fuel Properties
Aroraatics
Olefins
Saturates
Lead
Phosphorous
Sulfur
30 max.
15 max.
20 nin.
5 rain.
balance
less than 0.01 grams/gallon
less than 0.003 crans/^al
less than 200 ppm wt./v/t.
6. Conclusion.
The above test sequence is derived from similar
tests conducted in the automotive industry, specifically
the AC-823 test. Modifications have been made to test
converters more nearly in the conditions typical of
their use in cars built to the interim 1975 standards.
236
-------
9 PERFORMING ORGANIZATION NAME AND ADDRESS
Health Effects Research Laboratory
Office of Research & Development
U.S. Environmental Protection Agency
Research Triangle Park, N.C. 27711
NO
EPA-600/3-75-010d_
2.
I. TITLE AND SUBTITLE
ANNUAL CATALYST RESEARCH PROGRAM REPORT
Appendices, Volume III
5. RCPORT DATE
September 1975
6. PERFORMING ORGANIZATION CODE
AUTMORISI
0. PERFORMING ORGANIZATION HF O ''JO.
Criteria and Special Studies Office
T- ACCESSION>NO.
10 PROGRAM ELEMENT NO.
1AA002
11. CONTRACT/GRANT NO.
17. SPONSORING AGENCY NAME AND ADDRESS
Same as above
13 TYPE OF REPORT AND PERIOD COVLRID
Annual Program Status 1/74-9/7J4
14 SPONSORING AGENCY CODE
EPA-ORD
b. SUPPLEMENTARY NOTES
This is the Summary Report of a set (9 volumes plus Summary).
See EPA-600/3-75-010a thru OlOc, OlOe thru OlOj.Report to Congress.
16. ABSTRACT
This report constitutes the first Annual Report of the ORU Catalyst Research
Program required by the Administrator as noted in his testimony before the
Senate PUblic Works Committee on November 6, 1973. It includes all research
aspects of this broad multi-disciplinary program including: emissions charac-
terization, measurement method development, monitoring, fuels analysis,
toxicology, biology, epidemiology, human studies, and unregulated emissions
control options. Principal focus is upon catalyst-generated sulfuric acid
and noble metal particulate emissions.
17
KEY WORDS AND DOCUMCNT ANALYSIS
DESCRIPTORS
Catalytic converters
Sulfuric* acid
Desulfurization
Catalysts
Sul fates
Sulfur
Health
ll.IDENTIFIERS/OPEN ENDED TERMS
Automotive emissions
Unregulated automotive
emissions
Health effects (public)
<-. COSATI I lklil/(.u>ii|i
STATEMENT
Available to public
SECURITY CLASS (lltilHrporl/
21 NO OF PAGES
72 PRICE
Unclassified
------- |