SEPA
United States      Industrial Environmental Research  EPA-600/7-80-088
Environmental Protection  Laboratory          April 1980
Agency        Research Triangle Park NC 27711
A Research Plan to
Study Emissions from
Small Internal
Combustion Engines

Interagency
Energy/Environment
R&D  Program Report

-------
                  RESEARCH REPORTING SERIES


 Research reports of the Office of Research and Development. U.S. Environmental
 Protection Agency, have been grouped into nine series. These nine broad cate-
 gories were established to facilitate further development and application of en-
 vironmental technology. Elimination  of  traditional grouping was  consciously
 planned to foster technology transfer  and a maximum interface in related fields.
 The nine series are:

     1. Environmental Health Effects Research

     2. Environmental Protection Technology

     3. Ecological Research

     4. Environmental Monitoring

     5. Socioeconomic Environmental Studies

     6. Scientific and Technical Assessment Reports (STAR)

     7. Interagency Energy-Environment Research and Development

     8. "Special" Reports

    9. Miscellaneous Reports

This report has been assigned to the INTERAGENCY ENERGY-ENVIRONMENT
RESEARCH AND DEVELOPMENT series. Reports m this series result from the
effort funded under the 17-agency Federal Energy/Environment Research and
Development Program. These studies relate to EPA's mission to protect the public
health and welfare from adverse effects of pollutants associated with energy sys-
tems. The goal of the Program is to assure the  rapid  development of domestic
energy supplies in an environmentally-compatible manner by providing the nec-
essary environmental data and control technology. Investigations include analy-
ses of the transport of energy-related  pollutants  and their health and ecological
effects;  assessments  of, and development of, control technologies  for  energy
systems; and integrated assessments of a wide-range of energy-related environ-
mental issues.



                       EPA REVIEW NOTICE
This report has been reviewed by the participating Federal Agencies, and approved
for  publication. Approval does not signify that the contents necessarily reflect
the  views and policies of the Government, nor does mention of trade names or
commercial products  constitute endorsement or recommendation  for use.

This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia 22161.

-------
                                EPA-600/7-80-088

                                          April 1980
  A Research Plan  to Study
      Emissions from Small
Internal  Combustion  Engines
                      by

                 James W. Murrell

       Systems Research and Development Corporation
                  P.O. Box 12221
        Research Triangle Park, North Carolina 27709
               Contract No. 68-02-3113
             Program Element No. INE624A
           EPA Project Officer: John H. Wasser

        Industrial Environmental Research Laboratory
      Office of Environmental Engineering and Technology
            Research Triangle Park, NC 27711
                   Prepared for

        U.S. ENVIRONMENTAL PROTECTION AGENCY
           Office of Research and Development
               Washington, DC 20460

-------
                                  ABSTRACT

     This report examines some of the requirements for investigating
environmental status of small internal combustion engines.  These engines
range in size from 1 1/2 hp to 15 hp and power a variety of equipment by
home owners and industrial members.
     With the general growing concern in EPA of identifying sources of
potentially carcinogenic emissions, there exists a possibility that these
small internal combustion engines are a problem source.  Research to char-
acterize emissions from this source has largely been limited to critical
pollutants, even though the small internal combustion engine is an incom-
plete combustor.  It follows that some carcinogens and other hazardous
compounds are probable.
     The basic requirements addressed in this report include:
     a)   analytical  equipment
     b)   experimental systems design
     c)   statistical  experimental design
     Work on this document was performed under EPA Contract No. 68-02-3113
under the direction  of Mr.  J.H.  Wasser, Project Officer.
                                     ii

-------
                                  CONTENTS

Abstract 	 ..... ii
Figures	tv
Tables	lv
     1.  Introduction	1
     2.  Information Search and Assessment 	  3
     3.  Analytical Equipment  	  9
     4.  Experimental System Design  	 13
     5.  Statistical Experimental Design 	 19
     6.  Summary	30
References	31
Appendices
     A.  Annotated Bibliography  	  A-l
     B.  Analytical Equipment Characteristics  	  B-l
     C.  List of Engine Manufacturers/Engines  	  C-l
                                     iii

-------
                                   FIGURES
Number
                                                                        Pajge
  1          Source Assessment Sampling Training Schematic ....       15
  2          Typical Survey Sampling System  	       17
                                  TABLES
Number                                                                 Page
  1     Information Gathering Form for Analytical Equipment ....      10
  2     Subplot Formulation as a 2 x 3 Factorial   	 ....      21
  3     Emission of SIC Engines by Block (Mode x RPM)  and
          Treatment (CS x Age)	      21
  4     Anova Table for Model
  5     Anova Table for Model
  6     Test Matrix for Inexpensive Experiments ...  	      26
  7     Test Matrix for Expensive Experiments 	      27
                                   iv

-------
                                 SECTION 1
                                INTRODUCTION
     Over the past decade, population and energy consumption  by way of the
internal combustion engine has increased dramatically.   Concomitant with
these increases has been a dramatic increase in the nation's  air pollution
problem.  An elevated awareness of the hazardous aspects of emissions  from
these engines has .been developed.   Most of the research and the ensuing
legislation has been aimed at the gasoline powered automobile,  and rightfully
so.  Recently, diesel engines have been given greater attention because of
their ever increasing numbers.  However, relatively little attention has
been focused on the small internal combustion engines in spite of their wide
spread usage.  Here "small" is defined as 15 horsepower or less and includes
engines used to power equipment such as garden tractors, motor tillers, lawn
mowers, chain saws and other recreational, industrial and agricultural equip-
ment.  General growing concern to identify sources of potentially carcino-
genic emissions, has caused the EPA to explore the possibility that internal
combustion engines are a problem source.  These engines are incomplete
combustors, therefore, there is a high probability that carcinogens and
hazardous compounds are emitted.
     The purpose of designing a comprehensive research plan is to allow
emissions from these small internal combustion engines to be characterized.
This characterization will account for trie interactive impact of several

-------
factors on determining the acceptability of projected ambient concentrations
of various emission compounds.   As stated earlier, the primary factors are:
     -age of engine
     -carburetor setting
     -revolutions per minute
     -mode (or load)

-------
                                  SECTION 2
                      INFORMATION SEARCH AND ASSESSMENT

     The objective of this task is to find what is  known about the
environmental aspects of small  engines.   In performing this task, SRD team
members used several approaches.  Among these were:
     -Review of small engine/engine driven equipment
     -Review of literature using EPA on-line search system
     -Review of related materials in the libraries  of area universities
     -Interviews with local distributors
     In addition, requests for information pertaining to engine size, type
(# of strokes), application, emission data, fuel-air mixtures, life expect-
ancy, duty cycle, sales and usage patterns were made from nine major small
engine manufacturers.
     This task effort confirmed our suspicion of the paucity of information
related to distribution, emission characteristics and the health/ecological
effects of these small engines.  One dramatic indictment of this situation
was found in an EPA document containing c;uantitative information for eighty
source categories.  These categories were selected by EPA as those common to
many areas of the United States and would potentially benefit most from
application of control devices.  The source categories are classified into
eleven major areas with internal combustion being considered a major area.
The internal combustion engine category consisted of one sub-category, diesel
and dual fuel engines.  Not only were small internal combustion engines

-------
 omitted,  data  for  the one sub-category was too sketchy for the category to
 be  developed.   The internal  combustion category is the only category of the
 twelve where there exists an  information void.  The remainder of this section
 will  explore the potential  impact of some of these emissions.
 2.1   HEALTH
      Among the  more prominent suspected emission are:
      -hydrocarcons
      -carbon monoxide
      -oxides of nitrogen
      -particulates
      Some detrimental health effects of these compounds are well known.
 Unburned hydrocarbons have an objectionable odor, contribute to photochemical
 smog  and are possibly carcinogenic.  Hydrocarbons may also show up as partic-
ulate matter.   Studies have shown that high molecular weight hydrocarbons
 have  been carcinogenic in animals.
      The toxicity of carbon monoxide has been well documented.  It occurs
 because blood hemoglobin has a higher affinity for carbon monoxide than for
oxygen.  After  approximately one hour of exposure to carbon monoxide at 600
ppm,  humans go  into a coma.   Death usually occurs after one hour of exposure
at 800 ppm.  There is epidemiological data suggesting increasing incidence
of mortality from myocardial infarction after continual average weekly
exposure to carbon monoxide concentrations of 8 to 14 ppm.
     Oxides of nitrogen have a tendency to combine with lung moisture to
form dilute nitric acid.   This may cause respiratory problems over extended
periods of time.  Oxides of nitrogen are also known to settle on blood
hemoglobin.  Oxides of nitrogen may be regarded as respiratory pollutants.
One of the major groups of respiratory pollutants is that of pulmonary
                                     4

-------
irritants.  Many respiratory irritants  contribute  to  the  development of
cancer.  By interfering with ciliary activity and  retarding  the  flow of
mucous in the bronchi, they enhance retention of carcinogenic  particles  in
the lungs, and in this way, encourage tumor formation.
2.2  NOISE
     The past twenty years are characterized by the increased  concern  and
activity of citizens in environmental issues.  Noise pollution is  an  issue
with which both individual citizens and community  and environmental organiza-
tions have expressed discontent.  Most research in this area of pollution has
been placed on the response of the individual to airport noise and has
focused primarily on the physical/acoustical dimensions of noise exposure.
     It is well known that hearing damage can result from high noise  and
from overexposure to sounds that are at a lower level than the average lawn-
mower.  Sounds levels are usually measured in decibels.  Ordinarily,  speech
might register 60 decibels and sounds from a low flying aircraft may register
120 decibels or more.  Hearing discomfort begins at about 95 decibels and
pain begins at approximately 140.  Hearing damage  may begin at a much lower
level and is a function of exposure.
     Most existing technology can adequately monitor the noise pollution
resulting from different classes of small engines.
     The most cost effective study for examining the impact of hearing loss
due to small engines is a retrospective study.  This study would identify a
population of individuals who have known time of exposure to sounds from
certain types of small engines  (professional lumberjacks, etc).  These
individuals would be compared to a "control" group while controlling for
potential confounding variables.

-------
 2.3   ESTIMATING  NATIONAL HEALTH EFFECT
 2.3.1   Overall Effect
      The  estimation of national health effects requires information on
 several variables.  The variables include:
      -Distribution of Engines
      -Total population of Engines
      -Emission rates
      -Usage (exposure) rates
     These variables are largely unknown and impossible to determine.  It
would be  impossible to use gasoline sales or engine sales as crude proxy
measures.  In the former case, fuel sales from small engines can not be
distinguished from fuel sales for automotive equipment.  On the other hand,
engine sales do not reflect engine use nor do they reflect engine life.
     One might try estimating some of these variables purely on a deductive
basis.  This could be done by assuming a reasonable exposure time for the
"average" American (i.e., lawn mowing, wood sawing, etc.).  An overall
effect could then be estimated provided that sufficient data on constituent
emissions and their health impact exists.  This information would then be
combined with user population data to estimate a national health effect.
This kind of estimate is crude at best.  One study applied such a method in
an example aimed  at determining the annual exposure time of lawnmower users
applying the following assumptions:
     1)   Each residential  lawn cover 10,000 square feet
     2)   To account for commercial  usage (plants, schools, etc.) and
         sharing  among families, each mov/er cuts two lawn areas
     3)   Each mower cuts 15-inch swath after correcting for overlap

-------
     4)  Mower speed is 2 feet/second
     5)  Grass growing season is 80 days
     6)  Cutting interval during season is 10 days
     It is readily seen that these assumptions,  though reasonable,  are
subject to a large amount of variability.
     Ultimately, better estimates can be gotten  by  performing a stratified
random sampling procedure.  The scope of the sample should be national and
stratification variables could include geographical location, commercial
versus non-commercial applications, season and engine characteristics among
other potential factors.
2.3.2  The Individual Effect
     Section 2.1 examined a few of the known detrimental health effects that
particular emissions have on an individual.  These  effects are expected to
be a function of exposure time as well as the concentration of the various
components of mass emissions.  Once threshold dosages of particular compounds
are identified, the "main" effects for that compound may be estimated by
appropriate modifications to the empirical model subsequently described.
     For a particular source pollutant from a small engine, one could
approximate the dose that an operator receives by
                                        .-T
                               D = RA    C(t)dt
where R = Volumetric lung capacity.
      A = The body retention rate.
   C(t) = Concentration of emissions as a function time.
Under static conditions C(t) would remain constant so that the dosage could
be estimated by
                               D = RAC(tQ)T

-------
where C(t ) is the steady state concentration and T is the length of time
that the operation is exposed to the emission.
     It should be noted that this model does not account for interactive
effects and is useful only as a first level estimate of individual dosage.

-------
                                 SECTION  3
                            ANALYTICAL EQUIPMENT
     Since little is known about small engine emissions,  one  of the  basic
objectives of the experiment will be to test these emissions  to determine  the
classes of substances that are known or suspected to have adverse health and
environmental effects.  This can best be accomplished with a  Level  1
assessment utilizing an EPA-developed phased approach.
3.1  THE PHASED APPROACH
     The phased approach, as developed by the Process Measurements Branch
(PMB) of the Environmental Protection Agency, required three  separate levels
of sampling and analytical effort.  The first level, Level 1, utilized
quantitative sampling and analysis procedures that yield final analytical
results accurate to within a factor of 3 of the samples.   Level 1 is designed
to (a) provide preliminary environmental assessment data, (b) identify
problem areas, and (c) formulate the data needed for the prioritization of
energy and industrial processes, streams within a process, components within
a process, components within a stream, and classes of materials for further
consideration in the overall assessment.  The second sampling and analysis
effort, Level 2, is directed by Level 1 results and is designed to provide
additional information that will confirm and expand the information gathered
in Level 1.  This information will be used to define control  technology
needs, and may, in some cases, give the probable or exact cause of a given
problem.  The third phase, Level 3, involves monitoring the specific problems

-------
 identified in Level  2 to provide  information  for  control  device  design  and
 development.   For example,  if a Level  1  test  indicated  that  polycylic
 organic material  (POM)  might  be present  in  significant  amounts and  also
 gave a positive  mutagenicity  test,  Level  2  sampling  and analysis would  be
 designed to determine the exact quantities  of organic constituents, the
 percentage of POM, and  the  identity of as many specific POM  compounds present
 as  is  economically possible.   In  addition,  using  the Level 1 data and any
 available  Level  2 results,  the sample  would be retested for  cytotoxicity and
 mutagenicity  in order to confirm  and expand the total bioassay information,
 A test for carcinogenicity would  also  be run  if the results  of these test
 were positive.
     The phased approach offers potential benefits in terms  of the quality of
 information that is obtained  for  a  given  ,evel of effort and in terms of the
 costs  per  unit of information.  This approach  has been  investigated and
 compared to the more  traditional  approaches and has been found to offer the
 possibility of substantial savings  in  both  time and funds required for
 assessment.
 3.1.1  Level  1 in a Phased Approach
     The Level 1  sampling and analysis program is designed to produce a
 comprehensive survey of emissions  from any  industry or energy-generating
 facility that might be of environmental consequence.   This survey shows,
within broad general  limits, the absence or presence, the approximate
concentrations, and the emission rates of inorganic elements, selected
anions, and classes of organic compounds in gaseous,  liquid,  and solid
samples.  Any particulate matter suspended in  the effluent gases is  analyzed
separately for chemical  composition, for size, and for other  physical
parameters that can be determined  by microscopic  examination.  Selective
                                    10

-------
biotesting is performed on samples  to obtain  information  indicative  of  the
possible human health and ecological  effects  of the material.   If  it can  be
proven that equivalent Level  1  data exist for all  streams of interest,  then
a Level  1 effort need not be  conducted.   If only partial  data exists, then a
complete complement of Level  1  tests  must be  performed on all  streams.
     The area of analytical equipment is one  of rapidly developing technology.
It is important to know what  equipment is available and at what cost.  It is
important that analytical equipment used during Level  1 assessment provide a
good approximation to the true levels of the compounds being sampled.  It is
expected that IERL/RTP or the contractor who actually performs the experiment
has analytical equipment which is suitable for Level 1 assessment.  One of
the tasks of the contract under which this report was written was  to review
and compile some of the more cost-effective analytical equipment.   Table 1
depicts the elements of a taxonomy of desirable information characteristics
for selecting some of this equipment assuming that no purchases have yet
been made.  Appendix B contains a representative compilation of some of this
equipment.
                                    11

-------
Taxonomic Dimension
Sampling
Elements
Method
Volume
Maximum Temperature Input
Collection Efficiency
Performance
Accuracy
Reproducibility
Linearity
Noise
Lag Time
Retention Time
Fall Time
Zero Time
Span Drift
Operation
Ambient Temperature Range
Temperature Compensation
Relative Humidity Range
Procedure
Unattended Period
Maintenance
Requirements
Power
Weight
Dimensions
Features
Output
Training for Operation
Cost
   Table 1.  Information gathering form for analytical  Equipment.
                                    12

-------
                                 SECTION  4
                         EXPERIMENTAL SYSTEM DESIGN

     In selecting an experimental system, sampling should be  designed  to
ensure that the emissions obtained are representative of those  encountered
under normal operating conditions.  This  task is nearly impossible for small
internal combustion engines.   Large variability among usage patterns  and
conditions account for this difficulty.  Some of these high variability
variables include exposure time, proximity of operator to engine and  meteoro-
logical conditions.  These factors need not be a major concern  during initial
assessment since their impact may be evaluated by modeling.  The primary
concern during initial assessment is obtaining sufficient amounts of  the
various pollutants to adequately characterize the stream.  This requires  mak-
ing meaningful measurements.   The following list of general criteria  should
be considered when setting up the experimental system:
     A.  The nature of the emissions should be classified according to
         whether they are gases, liquids, particles, or some
         combination of the three.
     B.  Efforts should be made to ensure that the emission can be
         measured separately from other sources of emission.
     C.  Procedures should be implemented to ensure that sampling does
         not significantly alter the process.
     D.  Possible reactive effects between different emissions should
         be considered.
                                     13

-------
      E.   Transport  air should be controlled so that emission concentration
          is  maintained at a measurable level.
 4.1   SAMPLE  ACQUISITION
      Stationary source particulate matter sampling and analysis have been
 restricted to streams of high mass loading until recently, because the flow
 rates through sampling equipment had not been high enough to collect an
 adequate  amoung of  material in a reasonable length of time.
      Because of this restriction, the development and application of control
 technology, which requires effluent information on four particulate size
 ranges, has been hampered.  It has also limited health effects studies, which
 require information on the distribution and composition of respirable and
 nonrespirable particulate size classes, the presence of volatile organic
 compounds, and the  presence of trace elements to be complete.  To correct
 this situation, EPA(IERL-RTP) has developed and specified the use of the
 Source Assessment Sampling System (SASS)* for the collection of particulate
 samples and volatile matter from ducted emissions (Figure 1).
     The SASS train consists of a stainless steel probe that connects to
 three cyclones and  a filter in an oven module, a gas treatment section, and
 an impinger series.  Size fractionation is accomplished in the cyclone
 portion of the SASS train, which incorporates the three cyclones in series to
provide large collection capacities for particulate matter nominally size-
classified into three ranges:  (a) >10 ym, (b) 3 ym to 10 yin, and (c) 1 ym to
3 ym .  By means of a standard 142-mm filter, a fourth cut, >1 ym, is also
obtained.   The gas  treatment system follovs the oven unit and is composed of
four primary components:   the gas cooler, the sorbent trap, the aqueous
condensate collector, and a temperature controller.  Volatile organic
                                     14

-------
en
                                             CONVECTION
                                             OVEN
 ISOLATION
BALL VALVE

        FILTER
                                                                                       GAS COOLER
                  SS PROBE *  |	\-
            S-TYPE PITOT

               DRY GAS METER/ORIFICE METER
               IMP/COOLER
               TRACE ELEMENT
               COLLECTOR
                CENTRALIZED TEMPERATURE
                  AND PRESSURE READOUT
                     CONTROL MODULE
                                             SORBENT
                                             CARTRIDGE
                                                               CONDENSATE
                                                               COLLECTOR
                                                         TWO 1Oft3/min VACUUM PUMPS
                              'Figure  1.  Source Assessment Sampling Train Schematic."

-------
 material  is  collected  in  a  cartridge or  " ;rap" containing a sorbent, which is
 designated to  be  XAD-2, a microreticular resin with the capability of absorb-
 ing  a  broad  range  of organic species.  Volatile inorganic elements are
 collected in a  series  of  impingers that  follow the condenser and sorbent
 system.  The last  impinger  in the series contains silica gel for moisture
 removal.  Trapping of  some  inorganic species also may occur in the sorbent
 module.  The pumping capacity is supplied by two 10-ft /min, high-volume
 vacuum pumps, while required pressure, temperature, power, and flow condi-
 tions  are regulated through a main controller.  At least 60 A of power at 110
 V is needed for operating the sampling equipment.
 4.2  SAMPLING SYSTEM DESIGN
     The Quasi-stack method is recommended as a means of ducting total engine
 emissions to the SASS train.  The Quasi-stack Measurement System consists of
 an enclosure to capture the emission at  the source, an exhaust duct or stack
 in which the emissions are measured, and a blower or fan that directs the
 emissions through the measurement duct.  This method has been widely used for
 measuring industrial sources of fugitive emissions.  The methods can be used
 with small engine emissions since the basic methodological requirement is
 that emissions are isolable and capable of being enclosed.  Figure 2 shows a
 typical system.
 4.2.1  Hood Requirements
     Care must be taken to provide sufficient capture velocity at the hood
 opening.   This area may be computed from the relationship.
          Q = VA,
where     Q = air volume flow rate, cubic feet per minute
          V = air velocity, feet per minute
          A = hood face area, square feet (Figure 2)
                                    16

-------
                                                                                 Exhaust
                                                                                   f

(

-^ 	 »MJ mm. 	 ^\
\
d -!
1
Air flow
pitot
-
1
Part
p« 	 ja mm. 	 »•
Measurement
duct
icle
•
^
Gas
sampler ^
a

>*-
1
Jypa«
ir
Control
valve
15
Blow


/er

Source
                    "Figure  2.  Typical  Survey Sampling System."

-------
      In  order  to  effectively measure  the velocity, temperature and pressure
of  the flowing stream  to  determine the total flow rate and to provide the
most  efficient sample  flows, flow in  the measurement duct should be in the
turbulent  range with a Reynold's number of 2 x 10  for a typical smoothwalled
duct.  The Reynolds number  for air is roughly calculated as
                               Re = 110 cV
where Re = Reynolds number, dimensionless
       d = duct diameter, feet
       V = air velocity,  feet per minute
Since  V = Q/A
             2
and    A = -rrd  /4
by substitution,               Re = 140(Q/d)
   .     .   140  _  140  e
ana    a " Re~ " 2 x 10b " 0.0007Q
     The blower of fan used to provide the required air flow rate should, in
general, be selected to provide about twice the calculated rate to allow for
adjustments for inaccuracies in estimates or assumptions.  The actual flow
rate may be controlled by providing a variable bypass air duct downstream of
the measurement duct.   Actual  system layouts will, of course, be governed by
space requirements at the source site.  The minimum straight duct runs of
3 duct diameters upstream and downstream of the measurement and sampling
ports must be provided to ensure that the sampled flow reaches and remains
in fully developed turbulent flow with a uniform velocity profile.
                                    18

-------
                                  SECTION 5
                       STATISTICAL EXPERIMENTAL DESIGN
     In evaluating the results of any experiment, the inferences that can be
made are dependent upon the nature of the data.  It is quite possible to
sustain tremendous experimental costs while obtaining data from which no
inferences can be made.  Quite often researchers are forced to choose among
several methodological alternatives.  When using a factorial experiment, a
full factorial design is often the methodological choice.  In the case of an
experimental design for small engine emissions, one has four factors to
consider.  Given the proposed level of these factors, a full factorial model
would require the total number of runs to be:
     3 (Carburetor settings) x 3 (Models) x 2  (RPM) x 2  (Age) x 3 (Engines) =
     108 Test runs
     This number of runs is obviously costly and unreasonable.  The  following
sections will present alternate designs based  on the  following
recommendations:
     A)  Initial attention should be given to  3 to 4  horsepower four
         cycle engines since they represent the most  popular size in
         use.
     B)  Minimize the number of runs to be performed  on  the more
         expensive classes of emissions such as particulate phase
         polynuclear aromatic  (PNA) and vapor  phase PNA.
     C)  Perform replicate runs under "typical conditions"  for  the
         class of emissions  that are relatively  inexpensive to  perform.
                                      19

-------
      Two  plausible models will  be  present  for each  class  of emissions.  One
 model  utilizes  the split-plot  design  and is  a true  "experimental  design."
 The  second  model  utilizes a  linear model using weighted least  squares.
 5.1   THE  SPLIT-PLOT  DESIGN
      The  split-plot  design can  be  used  to  obtain  information about measures
 of emissions while reducing  the number  of  runs by the  incorporation of com-
 binations of two  factors.  These factors are divided into "whole  plots" and
 "subplots."  These two factors  may be depicted as Factor  One ("whole plot" or
 block  effects)  and Factor Two  ("subplot" or  treatment  effects) which contain
 I levels and J  levels respectively.  An important point here is that both
 factors may actually be made up of a combination  of other factors (i.e.,
 I =  Ia Ib).  As an example,  using  the factors in  the small engine experiment,
 Factor One may  consist of I = I  I. = 6 combinations where r = 3 (levels of
                               a   o                         a
 carburetion settings) and I.  = 2 (levels of engine age).   In this case,
 Factor One makes  up the whole plot design.  In this example, carburetor
 settings may be randomized among engines of different  age.  Factor two, the
 subplot factor would be divided similarly with J = Ja  Ju  = 6 being determined
 by 3 levels of mode randomized over two levels of RPM.  Assuming no signif-
 icant interaction between age and the other factors, the  blocks defined by
 mode and RPM form a (Table 2) 2 x 3 factorial.  The combination of Factor
 One and Factor Two then give rise to the 36 treatment  combinations of emis-
 sions depicted in Table 3.  The configuration assumes  that tests will be
 conducted using six engines.
5.1.1  Analysis Model
     Let   Yijklm  = " + 6i +  *J  + Ek + Yl +  Wjk  +  <0Y>jl +  kl

                                         +Eijkl             
-------
                             TABLE  2


               Subplot  Formulation as a 2 X3 Factorial


                                 MODE

                            100%  50%    0%
                 RPM
2600


3600
A
B
C
D
E
F
                             TABLE  3


Emission of SIC  Engines by Block (Mode xRPM) and Treatment (CS xAge)
^"^^v^Factor 1
Factor 2"*^^^
Mode: 100%
RPM: 2600
Mode: 100%
RPM: 3600
Mode : 50%
RPM: 2600
Mode : 50%
RPM: 3600
Mode : 0%
RPM: 2600
Mode : 0%
RPM: 3600
MR
1 2
EA1 EA2
EB1 EB2
EC1 EC2
ED1 ED2
F* F*
LE1 LE2
E* E*
tR Lp2
FR
3 4
EA3 EA4
EB3 EB4
EC3 EC4
ED3 ED4
E* E*
LE3 LE4
E* E*
bF3 LF4
F6
5 6
EA5 EA6
EB5 EB6
EC5 EC6
ED5 ED6
E* E*
LE5 hE6
E* E*
tF5 LF6
NOTES:


(1)  E..  where  1  =A, B.C. D,E, F   and  j =1, 2,3, 4,5, 6  are the
      ' J
     mean emissions for  cell   ij.


(2)  Cells containing *  are  quite  atypical  and would be omitted for

     the more expensive  runs.
                                   21

-------
 be the M-th observation (replicate at the 1-th level of RPM for the k-th
level of the mode, j-th level carburetor setting, i-th level of engine age.
The parameters of the model are defined as follows:
                      6. = Engine age effect
                      0. = Carburetor setting effect
                       J
                      E.  = Mode effect
                      Y1 = RPM effect
                  (0E)-k = Carburetor setting - mode interaction
                  (0Y)., = Carburetor setting - RPM interaction
                  (Ev)   = Mode - RPM interaction
                 (0EY)--k= Carburetor setting - mode - RPM interaction
This model  may be simplified when considering only the whole plot effect and
the subplot effects.   The model then becomes;
               Yijk=^ai + ej  +  <<*>1j-+B1k + E1Jkl  (Model2>
which represents the 1-th subplot at the j-th level of Factor Two ($..) within
the k-th block (whole plot) of the i-th level of Factor One (a^.  It  is
important to note that Factor Two is a combination of two factors (i.e.,
according to some specified randomized design scheme.  Here it is assumed
that treatment effect (Factor Two) are fixed and block effects (Factor One)
are random.  The estimates obtained are:
                    E[B.k:    =  0
                    v:Bik:    =  o
                                     22

-------
The estimates of the interactive and the fixed effects  are  obtained  by  ANOVA
procedures.   Tables 4 and 5 provide the appropriate ANOVA tables  for the  two
models.
5.2  ALTERNATE EXPERIMENT USING LINEAR MODELS
     Tables  6 and 7 depict alternate experiments which  may  be conducted using
six engines  for the "inexpensive" runs and four engines for the "expensive"
runs respectively.  Both of these experiments are examples  of "Incomplete
Block" designs and are most easily analyzed using a general linear models
approach.  One important assumption in the use of this  model is that the
combined effects of the independent variables (factors) are additive.  This
assumption seems fairly reasonable.  Subsequent sections describe the model
and some elementary matrix procedures which may be employed in estimation and
hypothesis testing.
5.2.1  Notations
     The general linear model is defined as:
                                Y = X3 + e
where X is an nxp matrix of rank p
-------
                      TABLE  4



               ANOVA TABLE FOR MODEL  I



  SOURCE                 DF                      EMS



WHOLE PLOT

                                              ?    9
  Age                    1                   a[




  CS                     2                   o!
                                              2    2
Whole plot error         2                   a£ + 6a~
SUBPLOT


  ...                    o                    Z ,  1 2 r 2
  Mode                   2                   aE   ~2~ ^ek




  RPM                    1                   aE + T ^2
  Mode * RPM             2                   a  +




  CS*RPM               2                   a|+|




  Mode * CS              4                   a  +
  Mode * CS * RPM        4                   a\ + |
  Subplot error
                          24

-------
                           TABLE  5

                   ANOVA TABLE  FOR  MODEL  II


      SOURCE                   DF                    EMS

WHOLE PLOT ANALYSIS            5
   (Between blocks)

                                                 2     2
   Factor one                  2                a^ +6aB +

                                                  2    °
   Whole plot error            3                 a£


SUBPLOT ANALYSIS               30
                                                 2
   Factor two                  5                a£ +

   Factor one -Factor two      10                 ~
      Interaction                               a.. +
   Subplot error               15                a_
 Where, DF = Degrees of Freedom
       EMS = Expected Mean Square
                                                 E       TO
                                                 2
                                25

-------
                                                 BLE 6
                                TEST MATRIX FOR INEXPENSIVE EXPERIMENTS
ro

Tpst
Run

1
2
3
4
5
6
7
8
9

Engine 1


rpm
2600
2600
2600
3ouu
3600
3600
3600
3600
3600


Mode
0
50
100
0
0
50
50
100
100


CS
FL
FR
MR
MR
FR
FL
MR
FL
FR
age =
= 0 years

Engine 2


rpm
2600
2600
2600
3600
3600
3600
3600
3600
3600


Mode
0
50
100
0
0
50
50
100
100


CS
MR
FL
FR
TL
FR
MR
FR
FL
MR

Engine 3


rpm
2600
2600
2600
3600
3600
3600
3600
3600
3600


Mode CS


Engine 4 Engine 5 Engine 6

0 FR same test same test same test
50 MR runs as run as run as
100 FL Engine 1 Engine 2 Engine 3
0 FL
0 MR
KQ FL
en CD
100 MR
100 FR


-------
                                           TABLE 7
                             TEST MATRIX FOR EXPENSIVE EXPERIMENTS
ro
age = 0 years
Test
Run
Engine 1
rpm
Mode a)
CS
Engine 2
rpm
?fede a)
CS
age = 5 years
Engine 3
Engine 4
1
2
3
4
2600
2600
3600
3600
50
100
50
100
MR
FR
FR
MR
2600
2600
3600
3600
50
100
50
100
FR
MR
MR
FR
same test
runs as
Engine 1
same test
run as
Engine 2

-------
                       v    II if carburetor setting = MR

                        2   [
                            0 otherwise
                       Y  _Jl if mode = 50%

                        3 ~\
                            10 otherwise
                            1 if mode = 100%


                            0 otherwise
                          Jl if RPM = 3600

                        C
                       ~D   (0 otherwise
                            [1 if AGE - 5 years


                       ~6   10 otherwise
     Thus each cell depicted in Tables 6 and 7 denotes an "observation"



of some compound.  In a strict sense, the X's represent those combination of



factors which resulted in the observation of a compound and a corresponding



level Y.  Thus, the matrix X will have the form:



                   X ~l_'» *M » Xo,  3J  4}  5*  6-J


This is a "main effects" model and does not include any terms for interaction.



Interaction terms may be created by simply multiplying the row elements of



the factors of interest.  As an example, an RPM—Age interaction is denoted



by Xc Xfi and if found by creating a new column whose rows reflect the
   *vD -s»O


product of the corresponding rows of X5 and Xg.
                                     28

-------
5.2.2  Estimation  of Parameters
     The unknown parameters  6  and e are estimated by the following matrix
relationships:
                            e  =  (x xrVv
                            S  =  Y'£I  -/(X'XrV Y/(n-p)
Let C be a given axp matrix  of rank a


-------
                                 SECTION 6
                                   SUMMARY
     This  report  is written  in a manner that will allow for methodological
 alternatives  during the actual implementation.  This will allow the implemen-
 tation  to  be  based on  the goal of  optimizing the study in terms of desirable
 criteria.  These  criteria may vary depending on numerous factors.  An example
 of  a highly possible criteria is cost-effectiveness.  This criteria may
 govern  the selection of certain analytical equipment.  As an example, in
 instances where there  is no  appreciable gain in performance, it is always
 most cost-effective to utilize equipment that is already on hand.  For this
 reason, the report reflects  some of the desirable characteristics of certain
 components of the experimental systems without making formal references to
 specific manufactures or their line products.
     The theme of flexible alternatives is carried over to the statistical
 design.  Two  designs are presented.  One statistical model is a "classical"
 experimental  design while the other is a linear models approach which readily
 accommodates  multiple replications.
     Current  small engine population data is sparse.  Getting this information
 for the estimate of national impact should be done by a carefully planned
 statistical sample.
     This report is the end-product of efforts to formulate a basic research
 strategy to study emissions from small internal  combustion engines.  This
work was performed under EPA Contract No.  68-02-3113.
                                    30

-------
                                 REFERENCES

 1.   Adams, J.   Selection and Evaluation of Sorbent Resins for the Collection
     of Organic  Compounds.  EPA-600/7-77-044, April 1977.

 2.   Bowen, Joshua  S., Hall, Robert  E.  Proceedings of the Third Stationary
     Source Combustion Symposium; Vol.  I Utility,  Industrial, Commercial, and
     Residential  Systems.   EPA-600/7-70-050a, February 1979.

 3.   Bowen, Joshua  S., Hall, Robert  E.  Proceedings of the Third Stationary
     Source Combustion Symposium; Vol.  II Advanced Processes and Special
     Topics.   EPA-600/7-79-050b,  February 1979.

 4.   Bowen, Joshua  S., Hall, Robert  E.  Proceedings of the Third Stationary
     Source Combustion Symposium; Vol.  Ill Stationary Engine and Industrial
     Process  Combustion  Systems.  EPA-600/7-79-050c, February 1979.

 5.   Bowen, Joshua  S., Hall, Robert  E.  Proceedings of the Third Stationary
     Source Combustion Symposium; Vol.  IV Fundamental Combustion Research and
     Environmental  Assessment.   EPA-600/7-79-050d, February 1979.

 6.   Gushing, K.  M.,  et  al.  Particulate Sampling  Support:  1977 Annual  Report.
     EPA-600/7-78-009, January  1978.

 7.   Dorsey,  S.  A.,  Johnson, L.  D.   Environmental  Assessment and Sampling:
     Phased Approach  and Techniques  for Level  1.   EPA-600/2-77-115,  June 1977.

 8.   Duke,  K. M., et al.  IERL-RTP  Procedures  Manual:  Level 1  Environmental
     Assessment  Biological  Tests for Pilot Studies.  EPA-600/7-77-043,  April
     1977.

 9.   Gallant, R.  F.,  et  al. Characterization  of Sorbent Resins for  Use in
     Environmental  Sampling.   March 1978.

10.   Hamersma, S. L., et al.   IERL-RTP Procedures  Manual:  Level  1  Environ-
     mental Assessment.   EPA-600/2-76-160a,  June 1976.  .

11.   Hare,  Charles T.,  Springer, Karl J.   Exhaust Emissions  from Uncontrolled
     Vehicles and Related Equipment Using  Internal Combustion  Engines:   Part 2,
     Outboard Motors.  APTD-1491, January  1973.

12.   Hare,  Charles T.,  Springer, Karl J.   Exhaust Emissions from Uncontrolled
     Vehicles and Related Equipment Using  Internal Combustion  Engines:   Part 4,
     Small  Air-Cooled Spark Ignition Utility Engines.   APTD-1493, May 1973.


                                      31

-------
13.  Jaye, Fredrick C.  Monitoring Instrumentation for the Measurement of
     Sulfur Dioxide In Stationary Source Emissions.  EPA-R2-73-163, February
     1973.

14.  Kalika, P. W., et al.  Development of Procedures for the Measurement of
     Fugitive Emissions.  EPA-600/2/76-284, December 1976.

15.  Kolnsberg, H.J.  Technical Manual for the Measurement of Fugitive
     Emissions:  Quasi-Stack Sampling Method for Industrial Fugitive Emissions,
     May 1976.

16.  Lentzen, D. E., Estes, E. D., Gutknecht, W. F.  IERL-RTP Procedures
     Manual:   Level  1  Environmental  Assessment, 2nd edition.   EPA-600/7-78-201
     October 1978.

17.  Lilienfeld, Pedro et al.   Design, Development, and Demonstration of a
     Fine Particulate  Measuring Device.  EPA-600/2-77-077, April 1977.

18.  McAlevy, Robert F. Ill, Cole, Richard B.  Nitric-Oxide Measurement in a
     Simulated Spark-Ignition  Engine.  APTD-1498, January 1973.

19.  Smith, E. M., Little, Arthur D.   Sensitized Fluorescence for the
     Detection of Polycyclic Aromatic Hydrocarbons.  September 1978.

20.  Springer, George.   Engine Emissions,  Chapter 1,  "Engine  Exhaust
     Emission."
                                     32

-------
                                  APPENDIX A-l
                             ANNOTATED BIBLIOGRAPHY
SOURCE:
A Study of Emissions From 1966 -  1972 Light Duty Vehicles  in  Los  Angeles  and
St. Louis, Prepared by Automotive Environmental  Systems,  Inc.,  August 1973
DESCRIPTION:
A comprehensive study of exhaust and evaporative emissions from light duty
vehicles was performed in Los Angeles, California; and St. Louis, Missouri, to
determine the contribution to atmospheric pollution by 1966 through 1972 model
year vehicles.  This study was part of a total effort to measure the emissions
from light duty vehicles in six cities.  Automotive Environmental Systems,  Inc.,
under contract to the EPA, performed exhaust emissions tests on 170 vehicles
in Los Angeles, California; and St. Louis, Missouri, and evaporative emissions
tests on twenty 1972 model year vehicles in Los Angeles, California.
MEASUREMENT:
The emissions tests determined the levels of hydrocarbons, carbon monoxide,
carbon dioxide and oxides of nitrogen exhaust emissions as well as hydrocarbon
evaporative emissions.
                                      A-l

-------
 SOURCE:
 Dorsey, James A.  et  al,  Environmental Assessment Sampling and Analysis:
 Phased Approach and  Techniques  for  Level  1, June 1977.
 DESCRIPTION:
 A  sampling and analytical approach  has been developed for conducting environmental
 source assessments of  the feed, product,  and waste streams associated with
 industrial and energy  processes.
 This document presents an overview  of:  the historical development of the
 strategy, the concepts employed, the measurement techniques applied, and the
 costs of program  implementation.
              Components of an environmental source assessment program:
     1.  A systematic evaluation of the physical, chemical and biological
         characteristics of all streams associated with a process;
     2.  Predictions of the probable effects of those streams on the
         environment;
     3.  Prioritization of those streams  relative to their individual
         hazard potential, and
     4.  Identification of any necessary  control technology programs.
                 Ultimate goals of environmental source assessment:
     1.  To ensure that the streams from  a given processing scheme will
         be environmentally acceptable, or
     2.  Ensure that adequate control technology either exists or can be
         developed.

                          Information effective strategies:
Two clearly distinct strategies which would satisfy the requirements for
                                          A-2

-------
comprehensive information are the direct and phased approaches.   In  a  direct
approach, all streams would be carefully sampled and the samples  subjected  to
complete detailed analysis for all detectable components at an overall  accuracy
of 50 percent.  In a phased approach, all streams would first be  surveyed using
simplified, generalized sampling and analytical  methods which would  permit their
ranking on a priority basis (Level 1) i.e., very hazardous streams would be
distinguished from those less hazardous or relatively innocuous in nature.
Detailed sampling analysis (Level 2) would then be applied first to  streams
ranked in the highest priority by the Level 1 survey, and other streams would
be addressed in descending order to potential hazard.  Another phase, initiated
after consideration of Level 1 and 2 results, would involve the continuous
monitoring of "key" indicator materials to evaluate long-term process variability
(Level 3).

                  Cost-effectiveness--Direci and Phase Approaches:
Studies were conducted by the staff of the process measurements branch of EPA's
IERL-RTP with the objective of comparing the costs of direct and  phased  (elim-
inating low  priority streams) sampling and analysis approach.  In both cases
studied, the phased approach was  found to  be more  cost-effective  than the direct
approach.
                       Sampling  Programs in  a Phased Approach:
The most cost-effective  approach  clearly is  one  in which  detailed sampling is
performed  only on those  streams  demonstrated to  be potentially hazardous.  It
is not sound  practice  to attempt  to  define a detailed  sampling program until:
      1.  The general characteristics of  the  stream in  question have
         been evaluated.
      2.  The nature of any  unfavorable  sampling  system/sample interactions
                                         A-3

-------
         has been considered (e.g., chemical reaction, volatility loss).
MEASUREMENT:
     Level  (1) Analysis
     Level  (1) sampling provides a single set of samples acquired to represent
     the average composition of each stream.  This sample set is separated into
     solid, liquid, and gas-phase components.  Each fraction is evaluated with
     survey techniques which define its basic physical,  chemical, biological
     characteristics.

     In Level  (1),  the analytical  techniques and instrumentation have been kept
     as simple as possible to provide an effective level  of information at
     minimum cost.   Physical  analysis of solid samples is incorporated into
     Level  (1) because the size and shape of the particles  have a major effect
     on their  behavior in  process  streams,  control  equipment,  atmospheric
     dispersion,  and the respiratory system.   Some materials have characteristic
     physical  forms which  can aid  in their identification.

     Chemical  analyses to  determine the types of substances present are
     incorporated to provide  information for predicting:   control  approaches,
     atmospherical  dispersion/transformation, and potential  toxicity of the
     stream.   Biological assay  techniques are incorporated  as  a measure of the
     potential  toxicity.

     Level  (2)  Analysis
     Level  (2)  analyses  must  positively identify the materials  in sources which
     have already been defined  as  causing adverse environmental  effects.   These
     are the most critical  of all  three levels.   Each  sample in Level  (2)  assess-
     ing will  require  the  analyst  to select appropriate  techniques  based  on the
                                        A-4

-------
     information  developed  in Level  (1) and  the  information requirements of
     the  assessments.
     Level  (3)  Analysis
     The  analytical  procedures  for Level  (3) are specified  to  the  stream
     components being  monitored,  and it  is  ;;ot possible  to  define  the  exact
     form they may take.   These analyses  are oriented  toward  the time  variation
     in the concentrations of  key indicator materials.   Both  manual  and instru-
     mental techniques may be  used, providing they can be implemented  at the
     process site.  At Level  (3), continuous monitors  for selected pollutants
     should be incorporated in the analysis program as an aid in  interpreting
     the  data acquired through manual techniques.

METHODOLOGY:
     Level (1) Methodology and Components
     Level (1), the principal  subject of this section, is structured to produce
     a cost-effective information base for prioritization of streams and for
     planning any subsequent programs.  It seeks to provide input data to
     support evaluation of the following questions:
     --  Do streams leaving the  processing  unit have a finite probability of
         exceeding existing or future air, water, or solid waste  standards
         or criteria?
     --  Do any of the streams leaving the  processing unit contain any classes
         of substances that are  known or suspected  to have adverse environmental
         effects?
     --   Into  what general categories (classes)  do  these adverse  substances
         fall?
     --  What  are the most probable sources of  these  substances?
                                         A-5

-------
--  Based on the adverse effects and mass output rates, what is the
    priority ranking of streams?
--  For streams exhibiting potential environmental effects, what is the
    basic direction that control strategies are likely to follow?

The Level (1) measurement program provides information on the physical
characteristics, chemical composition, and biological effects of a given
stream.
                                 A-6

-------
SOURCE:
Duke, K. M., et al,  IERL-RTP Procedures Manual:   Level  1  Environmental
Assessment Biological  Tests for Pilot Studies, April  1977.
DESCRIPTION:
This manual focuses  on the Level 1  sampling and bioassy effort of the three
phased approach to performing an environmental source assessment — the testing
of feed and waste streams associated with industrial  processes in order to
define control technology need.  Each phase involves distinctly different sampl-
ing and analytical activities.

The three phased sampling and analytical strategy was developed to focus available
resources (both manpower and dollars) on emissions which have a high potential
for causing measurable health or ecological effects, and to provide chemical and
biological  information on all sources of industrial emissions.

The Level 1 sampling and analysis goal  is to  identify the  pollution potential
of a source.  Level 1 has at  its most  important  function the  selection,  in  order
of relative toxicity, of specific streams and components for  the  Level 2 effort.

The manual  presents the strategy of  the phased approach.   It  also presents  the
basic sampling procedures and the Level 1 protocol for the biological  tests
used to analyze the samples.   It briefly discusses possible bioassay procedures
for  Level 2 and 3.
                                         A-7

-------
SOURCE:
Jay, Frederic C., Monitoring Instrumentation for the Measurement of Sulfur
Dioxide in Stationary Source Emissions, February 1973.
DESCRIPTION:
     Substances:
    Experiments:
    Conclusions:
    Experiments:
The S02 stack gas monitoring instrumentation is comprised of
a sampling system, including any necessary sample condition-
ing equipment, and a measuring system including an analyzer
and recorder.  The system is to be used to continuously
determine quantitatively the concentration of S02 gas in exit
gas from stationary power plants using fossil fuels.
Five nondispersive infrared (NIDR) SOp monitors were studied.
The units considered were:  1) Leeds and Northrop - IR gas
analyzer, 2) Bendix Unor, 3) Intertech Uras, 4) Beckman Model
315A, 5) MSA - Lira Model 300.  All units used as their
principle means of detecting SOg the selective absorption of
infrared radiation by SOp molecules, and consisted of an IR
source, a chopper, a sample chamber, a reference chamber,
and a detector unit.
All  the instruments studies were only conditionally suitable
for use as continuous monitors since all suffered interference
from particulate matter, water vapor and vibration.
It was concluded that the proper ranking of the instruments
could only be done after the instruments had been subjected to
laboratory tests to measure their performance.
Commercially available SO^ analyzers based on the use of
electrochemical  principles were evaluated for use in
continuous monitoring S02 in stack gases.   The evaluation
                     A-8

-------
               included  a  review of:   basic  electrochemical principles
               involved, theoretical  and  practical  limitations,  possible
               interference from other stack gas  constituents, operating
               procedures, maintainability,  and  physical  construction.
               The instruments were of three types:   conductimetric,
               coulometric, and "fuel cell."
Conclusions:    Conductimetric instruments measure the change  in  conductivity
               of a solution resulting from the addition of S(L.  Instuments
               found to be capable of monitoring SCL in stack gases were:
               Mikrogas -  MSK - SO,, - El  by Calibrated Instruments Model  70
               Stack Monitoring System by Scientific Industries.
               Coulometric instruments are based on the principle of coulo-
               metric titration where electrogenerated halogen  (bromine or
               iodine) serves as the titrant.  Instruments recommended for
               possible use in stack-gas monitoring are:  Model  286/ or 400
               by  ITT Barton, Diffusion Stack Monitor Model 906A by Beckman
               Instruments.
               The "fuel  cell" - type  instruments are based on  a completely
               sealed sensor  functioning as  an electrochemical  transducer.
               Some of  the drawbacks of  this type of instrument are:
               1)  Membrane is not  highly specific to SO,,  (NO,, is a major
               interfering species);  and 2)  Membrane surface must  be kept
               free of  condensate  and  particulates.  Only the basis of data
               from the field test program,  the  following equipment is
               recommended for use in  SO^  control  technology development
               programs:
                    a.  Calibrated Instruments  Co.  Model  MSK-S02-E1
                                      A-9

-------
         conductometric SOp analyzer installed according to
         schematic 521-S1/2.
     b.  DuPont model 460 untraviolet SCL analyzer when
         calibration gas standards are used instead of the
         optical calibration filter for calibration purposes.
     c.  Intertech model URAS-2 non-dispersive infrared S(L
         analyzer when used with model  7651 probe and filter,
         model 7865 auto zero and calibration units.  The CMR
         5869 unit shall be set at 8 hour intervals.
These recommendations are made on the basis of:
     a.  Proven reliability and performance.
     b.  The availability of the unit as a complete system.
The conclusions from interpretation of the data concerning
the definition of the state-of-the-art in S02 monitoring are:
     a.  There is equipment available which is adequate for
         the monitoring purpose required by the New Source
         Performance Standards of December 23, 1971.
     b.  The equipment which does operate properly is that
         which is available as a complete system consisting
         of probe,  particulate filtration, sample conditioning
         and analyzer equipment.
     c.  No particular measurement technique is superior to
         all others.
     d.  A major operational problem encountered with most of
         the units  was the variance of  instability of the
         "zero" setting.
                    A-10

-------
SOURCE:
Ballant, R. R.  et al,  Characterization of Sorbent Resins for use in
Environmental Sampling, March 1978.

DESCRIPTION:
This technical  report has information pertaining to the use of chromatographic
techniques to characterize resins which are used to trap vapors in environmental
sampling schemes.  Two chromatographic techniques are described, frontal and
elution analysis.

Three diverse adsorbate groups, consisting of eight distinct chemical classes,
were studied as potential pollutants.

Sorbent modules are frequently employed as one of a number of collection devices
or stages  in a multi-purpose sampling device, such as the EPA-SASS train.  The
SASS train sorbent trap is primarily designed to capture organic species that
have sufficient volatility to pass through particulate filters upstream from
the sorbent bed.

One of  the more common methods of characterizing absorbents  is the use of gas
chromatography.  Characteristics data may be obtained by both elution analysis
methods.   The elution method introduced a small quantity of  sorbate  to the
sorbent in a short time.  In the frontal method the sorbent  is continuously
challenged with a steady  state concentration sorbate.

The data presented in  this report support the use of chromatographic elution
data to characterize breakthrough and  sorption capacity of  sorbent cartridges
containing synthetic resins.
                                        A-ll

-------
SOURCE:
Hamersma, S. L., et al, IERL-RTP, Procedures Manual:  Level 1 Environmental
Assessment, June 1976.
DESCRIPTION:
This manual describes a set of sampling and analytical procedures compatible with
the information requirements of a comprehensive Level 1 environmental assessment.
The phased sampling and analytical strategy was developed to focus available
resources (manpower and dollars) on emissions which have a high potential for
causing measurable or ecological effects, and to provide comprehensive chemical
and biological information on all sources of industrial emissions.

The phased approach requires three separate levels of sampling and analytical
effort.  The first level utilized quantitative sampling and analysis procedures
accurate within a factor of 2 to 3 and:  1) provides preliminary environmental
assessment data, 2) identifies problem areas, and 3) formulates the data needed
for the prioritization of energy and industrial processes, streams within a
process, components within a stream, and classes of materials for further
consideration in the overall assessment.
The manual is divided into two major sections:  sampling procedures and
analytical procedures.  The sampling section is divided into five chapters:
fugitive emissions, gases, aerosols, liquids, (including slurries), and solids.
The analytical section is divided into three chapters:  inorganic, organic and
bioassays.
                                        A-12

-------
SOURCE:
Hare, Charles T., Springer, Karl  J., Exhaust Emissions from Uncontrolled
Vehicles and Related Equipment Using Internal Combustion Engines:   Part 4
Small Air-Cooled Spark Ignition Utility Engines, APTD-1493, May 1972.

DESCRIPTION:
This document is Part 4 of the Final Report on Exhaust Emissions of
Uncontrolled Vehicles and Related Equipment Using Internal Combustion Engines.
Exhaust emissions from five gasoline-fueled, air-cooled utility engines were
measured using two types of steady-state procedures, measurements were taken
during transient operation.
The  study includes test data, documentation, and discussion on detailed emissions
characterization of five engines (one 2-strokes and four 4-stroke), as well as
estimated emission factors and national emissions impact.
MEASUREMENT:
The  exhaust  products measured during the emissions  tests  included:
      1.  Total hydrocarbons by FIA
      2.  Hydrocarbons, CO, C02 and  NO  by NDIR
      3.  02  by electrochemical analysis
      4.  Light hydrocarbons by gas  chromatograph
      5.  Total aliphatic aldehydes  (RCHO)  and  formaldehyde (HCHO)  by  the MBTH
         and chromotropic  acid method,  respectively.
                                         A-13

-------
SOURCE:
Hare, Charles T., Springer, Karl J., Exhaust from Uncontrolled Vehicles and
Related Equipment Using Internal Combustion Engines:  Part 7-Snowmobiles,
April 1974.

DESCRIPTION:
This is part 7 of the Final Report on Exhaust Emissions from Uncontrolled
Vehicles and Related Equipment Using Internal Combustion Engines.

Exhaust emissions from four snowmobile engines were measured using steady-state
"mapping" procedures, employing 29 combinations of speed and load for each
engine.
The engines tested were an Artie 440, a Polaris 335, a Rotax 248, and an OMC
528 rotary.  The first three engines listed are all 2-stroke vertical twins
with blower cooling, and the last engine is a blower-(and charge-) cooled rotary
combustion (Wankel) engine.

MEASUREMENT:
The gaseous exhaust constituents measured on a continuous basis during all the
test modes included:
     1.  Total  hydrocarbons by FIA:
     2.  CO, C02, NO, and HC by NDIR;
     3.  NO and NO  by chemiluminescence;
                  /\
     4.  Op by  electrochemical analysis
                                       A-14

-------
SOURCE:

Hare, Charles T., Springer,  Karl  J.  Exhaust Emissions from Uncontrolled Vehicles
and Related Equipment Using  Internal Combustion Engines:   Part 5 Heavy-Duty
Farm, Construction, and Industrial  Engines, October 1973.


DESCRIPTION:

This report is Part 5 of the Final  Report on Exhaust Emissions from Uncontrolled

Vehicles and Related Equipment Using Internal Combustion Engines.  The engine

categories covered in this report are heavy-duty gasoline and diesel engines

used in farm, construction,  and industrial applications.  Exhaust emissions from

twelve engines were measured, including eight diesels and four gasoline engines.


The  program of research on which this report is based was initiated by the EPA

to  (1) characterize emissions from a broad range of  internal combustion engines

in  order to accurately set priorities for  future control as required,  and

(2)  aid in  the development of more  inclusive national and regional  air pollution
                              '»
inventories.


MEASUREMENT:

The emissions to be measured for all the  engines included:

      1.  Hydrocarbons  by  FIA;

      2.  CO,  C02 and NO by  NDIR;

      3.  NO and  NO  by chemiluminescence;
                   /\

      4.  02 by electrochemical analysis-

      5.  Light hydrocarbons by gas  chromatograph;

      6.  Aldehydes by  wet chemistry;

      7.  Particulate by gravimetric analysis.   In  addition, hydrocarbons were

          to be measured by  NDIR  for gasoline engines, and smoke by the PHS

          full-time smoke-meter  for diesel engines.
                                        A-15

-------
SOURCE:
Hare, Charles T., Springer, Karl J., Exhaust Emissions from Uncontrolled
Vehicles and Related Equipment Using Internal Combustion Engines:  Part 6-
Gas Turbine Electric Utility Power Plants.
DESCRIPTION:
This document is Part 6 of the Final Report on Exhaust Emissions from Uncontrolled
Vehicles and Related Equipment Using Internal Combustion Engines.  In contrast
to the other phase of the subject contract, no measurements of emissions from
the source under consideration (Gas Turbine Electric Utility Powerplants) were
taken as part of the research project.   The reasons for this departure were
that information on gas turbine emissions available in the literature was deemed
sufficient (at least on the major emissions) and that the small test effort
possible within the scope of the contract would hardly add anything worthwhile
to that body of knowledge.
MEASUREMENT:
Emissions data include, NO, N02> and N0x measured by a variety of techniques;
a less substantial  amount of CO and hydrocarbon data; either CO,, or 02
(occasionally both)  for a given test; and scattered information on SO
                                                                     J\
particulate,  visible smoke, and less important pollutants.
                                       A-16

-------
SOURCE:
Hare, Charles T., Springer,  Karl  J.,  Exhaust Emissions from Uncontrolled
Vehicles and Related Equipment Using  Internal  Combustion Engines:   Part 3-
Motorcycle, March 1973.

DESCRIPTION:
This report is Part 3 of the Final  Report on Exhaust Emissions from Uncontrolled
Vehicles and Related Equipment Using  Internal  Combustion Engines.   Exhaust
emissions from seven motorcycles were measured using three separate procedures
for each bike.  Though motorcycles are currently exempt from Federal emissions
regulations, two of the procedures used for testing were based on those specified
in federal law for automobiles.
MEASUREMENT:
The first procedure used for the motorcycle tests was the Federal "7-mode" direct
sampling procedure (applicable to 1970 and 1971 model year light duty vehicles),
modified when necessary.  The exhaust components measured for the 7-mode tests
included hydrocarbons, CO, C02> and NO, all by NDIR.

The Federal  "LA-4" bag sampling procedure was also used  to test the motorcycles;
this procedure was modified as necessary.  Currently  it  specifies measurement
of the following:
     1.  Hydrocarbons by FIA
     2.  CO  and  C02 by NDIR
     3.  NO  and  NO  by chemiluminescence
                  /\
The final  procedure was a series of steady-state conditions  designed  to cover
the  range  of operating conditions experienced  by each motorcycle.   Exhaust
products measured during this  procedure  included:
                                        A-17

-------
1.  Total hydrocarbons by FIA;



2.  Light hydrocarbons by gas chromatograph (2 of the 7 machines only);



3.  Hydrocarbons, (2 of the 7 machines only);



4.  CO, C02, and NO by NDIR;



5.  NO and NO  by chemiluminescence;
             /\


6.  Op by electrochemical analysis;



7.  Total aliphatic aldehydes (RCHO) and formaldehyde (HCHO) by the MBTH



    and chromotropic acid method, respectively;



8.  Particulate by an experimental dilution type sampling device;



9.  Exhaust smoke (2-stroke machines only) using PHS full-flow smoke meter
                                  A-18

-------
SOURCE:

Hare, Charles T., Springer,  Karl  J.,  Exhaust Emissions from Uncontrolled
Vehicles and Related Equipment Using  Internal  Combustion Engines:   Part 2-
Qutboard Motors, APTD-1491,  January 1973.


DESCRIPTION:


This report is Part 2 of the Final  Report on Exhaust Emissions of Uncontrolled

Vehicles and Related Equipment Using  Internal  Combustion Engines.  The study

includes documentation and discussion on characterization of exhaust emissions

from four water-cooled 2-stroke outboard motors (Section III and IV), and

estimation of emission factors and national impact (Section V).


MEASUREMENT:

Exhaust emissions from four 2-stroke outboard motors were measured.


The components measured were:

     1.  Total hydrocarbons by FIA

     2.  CO, C02, NO, and hydrocarbons by NDIR

     3.  NO and NO  by chemiluminescence
                  A

     4.  Q£ by electrochemical analysis

     5.  Light hydrocarbons by gas chromatograph

     6.  Total aliphatic aldehydes and formaldehyde by  the MBTH and  chromatropic

         acid methods, respectively.
                                        A-19

-------
SOURCE:
Hare, Charles T., Springer, Karl J., Exhaust Emissions from Uncontrolled
Vehicles and Related Equipment Using Internal Combustion Engines:  Part 1-
Locomotive Diesel Engines and Marine Counter Parts, October 1972.

DESCRIPTION:
This document is Part 1 of the Final Report on Exhaust Emission from Uncontrolled
Vehicles and Related Equipment Using Internal Combustion Engines.
The program of research on which this report is based was initiated by the
Environmental Protection Agency to (1) characterize emissions from a broad range
of internal combustion engines in order to accurately set priorities for future
control, as required, and (2) assist in developing more inclusive national and
regional air pollution inventories.  This document, which is Part 1 of what is
planned to be a seven-part final report, concerns emissions from locomotive diesel
engines (and their marine counterparts) and the national impact of these engines.

The primary objectives of the locomotive portion of this project were to collect
useful emissions data on three locomotive diesel engines, and to use these data
in conjunction with supplementary data on emissions, number of units in service,
and annual  usage to estimate emission factors and national  impact.
MEASUREMENT:
The emissions to be characterized included:
     1.   Total  hydrocarbons;
     2.   Light hydrocarbons;
     3.   Aldehydes;
     4.   CO,  C02 NO by NDIR and chemiluminescence;
                                       A-20

-------
5.   NO  by chemiluminescence;
      /\


6.   02



7.   Smoke by a modified PHS opacity meter;



8.   Participate by an experimental dilition-type sampling system
                                   A-21

-------
 SOURCE:
 Kolnsberg, H. J., Technical Manual for the Measurement of Fugitive Emissions:
 Quasi-Slack  Sampling Method for  Industrial Fugitive Emissions, May 1976.

 DESCRIPTION:
 The objective of this manual is  to present the fundamental considerations
 required for the use of the Quasi-Stack sampling method in the measurement of
 fugitive emissions.
 EMISSIONS- Pollutants emitted into the ambient air from an industrial plant or
 other site generally fall into one of two types.  The first type is released into
 the air through stacks or similar devices designed to direct and control the flow
 of the emissions, and they may be readily measured by universally-recognized
 standard sampling techniques.  The second type is released into the air without
 control of flow or direction; these cannot be measured using existing standard
 techniques.
                          Categories of Fugitive Emissions
 A useful approach toward categorization of fugitive emissions is to group them
 according to the methods of their measurement.  Three basic methods exist—quasi-
 stack sampling, roof monitor sampling, and upwind-down-wind sampling.

 SAMPLING:
                             QUASI-STACK SAMPLING METHOD
This method has the fugitive emissions captured in a temporarily installed hood
or enclosure and vented to an exhaust duct or stack of regular cross-sectional
area.   The emissions are then measured in the exhaust duct using standard stack
sampling or similar well recognized methods.
Sources of fugitive emissions measurable by the quasi-stack method include:
                                       A-22

-------
    --Material  transfer operations
         Solids-conveyor belts, loading
         Liquids-spray, vapors
    --Process leaks
         Solids-pressurized ducts
         Liquids-pumps, valves
    --Evaporation
         Cleaning  fluids-degreasers, wash tanks
         Paint  solvent vapors-spray booths, conveyors
    --Fabricating  operations
         Solids-grinding,  polishing
         Gases-welding, plating

                             ROOF MONITOR SAMPLING METHOD
This method measures the fugitive emissions  entering the ambient air from
building or other enclosure openings  such as roof monitors, doors, and windows
from enclosed sources too numerous or unwie'dy to permit the installation of
temporary hooding.   Sampling generally is limited to a mixture of all uncontrolled
emission sources within the enclosure and requires the ability to make low air
velocity measurements  and mass balances of small quantities of material across
the surface of  the  openings.
                           UPWIND-DOWNWIND SAMPLING METHOD
This method  is  used to measure fugitive  emissions  from  sources  typically
covering large  areas  that  cannot be temporarily  hooded  and  are  not enclosed
in a  structure  allowing the  use  of the roof monitor method.
The upwind-downwind method quantifies  emissions  from  sources  such as material
handling and storage operations, waste dumps  and industrial  processes in which
                                        A-23

-------
 the  emissions  are  spread over  large areas or are periodic in nature.
 The  emissions  from such sources are quantified as  the difference between the
 pollutant concentrations measured  in the ambient air approaching (upwind) and
 leaving  (downwind)  the source  site.
 MEASUREMENT:
                                   Quasi-Stack Method
 Effective use  of the quasi-stack method requires that the source of emissions
 be isolable and that an enclosure  can be installed capable of capturing
 emissions without  interference with plant operations.
 The  quasi-stack method is mainly restricted to a single source and must be
 limited  to two or  three small sources that can be enclosed to duct their total
 emissions to a single sampling point.  In some cases, enclosing a portion of a
 process  in order to capture its emissions can alter that portion of the process
 by changing its temperature profile or affecting flow rates.  Emissions may be
 similarly altered  by reaction with components of the ambient air drawn into the
 sampling ducts.
 The  quasi-stack method is useful for all types of emissions.  One of the reasons
 for  this is it will provide measurable samples in generally short sampling times
 since it captures essentially all of the emissions.
                                Roof Monitor Method
 The  roof monitor method requires that the source of emissions be enclosed in a
 structure with a limited number of openings to the atmosphere.
 This method is usually dependent on or influenced by gravity in the transmission
 of emissions, and it may not be useful  for the measurement of larger particles
 which may settle within the enclosure being sampled.  Emission generation rates
must be high enough to provide pollutant concentrations of measurable magnitude
                                        A-24

-------
after dilution in the enclosed volume of  the  structure.
                               Upwind-Downwind  Method
This method is generally used where neither of  the other methods  would be
successful.  The method is strongly influenced  by meteorological  conditions,
and requires a wind consistent in direction and velocity throughout the sampling
period as well as conditions of temperature,  humidity and ground  moisture
representative of normal ambient conditions.
OTHER CONSIDERATIONS:
                                 Sampling Strategies
Fugitive emissions measurements can  be separated into two classes.  Survey
measurement systems, designed  to screen emissions and provide gross measurements
at  a relatively  low  level of  effort  in time and  cost.   Detailed  systems  are
designed to isolate, identify and  quantify constituents with increased accuracy
and higher investments  in time and cost.

                                   Test  Strategies
Approaches that may  be taken to successfully complete  a testing  program  using
 the quasi-stack sampling method are  described.
                         Instrumentati on/Equi pment/Faclli ti es
 A description of the instrumentation arrays  to be used to collect the samples
 and meteorological data identified in the approach description.
 A description of the facilities required to operate the measurement program,
 including work space, electrical power,  support from plant personnel, special
 construction, etc.
                                   Quality Assurance
 The test  plan should address  the development of a quality assurance  program.
 This QA program should  be an  integral part of the measurement program and be
                                        A-25

-------
 incorporated  as  a  portion  of  the  test  plan  either directly or by reference.
                            Quasi-Stack Sampling Strategies
 This method  is used  to quantify the emissions from a source by capturing the
 emissions, entrained  in  the ambient air,  in a temporary hood or enclosure built
 over or  around the source  and  directing the captured stream through a duct of
 regular  cross section for  measurement,  sampling and analysis using standard
 stack  techniques.
                         Survey Quasi-Stack Sampling Strategy
 A  survey measurement system is designed to  provide gross measurements of
 emissions to determine whether any pollutant constituents should be considered
 for more detailed  investigation.  A quasi-stack measurement system consists
 basically of a hood or other enclosure to capture the emissions at the source,
 an exhaust duct or stack in which the  emissions are measured, a fan or blower
 to direct the emissions  through the measurement duct, and the emissions sampling
 equipment.
 SAMPLING EQUIPMENT:
 Particulate pollutants may  be grossly measured conveniently using any of a
 large variety of filter  impaction devices.  Gaseous pollutants may be grab-
 sampled  for laboratory analysis into suitably-sized vessels added to the
 particulate sampling train  or separate sampling ports elsewhere in the
measurement duct.
An alternative method for the measurement of particulates and volatile matter
 is the recently developed source assessment sampling system (SASS) train.  The
 train consists of a stainless steel probe that delivers the sample to an oven
module.  This device, used  in combination with a gas-sampling, provides all
 the information required as the nature and  composition of the pollutants in
                                        A-26

-------
                                 Sampling Techniques
The primary concern of the sampling design is that sufficient amounts  of the
various pollutants are collected to provide meaningful  measurements.
Information on detailed quasi-stack sampling strategy,  sample equipment and
design and sampling techniques are described.

                                  Quality Assurance
The basic reason for quality assurance on a measurement program is to insure
that the validity of the data collected can be verified.
An Appendix is also furnished in this document.  It represents an application
of the quasi-stack fugitive emissions measurement system selection and designed
criteria to a gray-iron foundry mold pouring operation.
                                        A-27

-------
 SOURCE:
 Lear, C. W., Charged Droplet Scrubber For Fine Particle Control:  Laboratory
 Study, September 1976.

 DESCRIPTION:
 This document gives results of a feasibility study of the application of
 charged droplet scrubbing for fine particle control.  Results, using the TRW
 charged droplet scrubber, indicated that the method is feasible and applicable
 over a wide range of conditions.  In the charged droplet scrubber the electrical
 interaction mechanisms exist in addition to the normal impact and diffusional
 scrubbing mechanisms.  Electrical interaction is strong in the 0.1 to 1.0
 micron particulate size range where the normal  mechanisms lack effectiveness.
 Collection efficiencies as high as 80% for 0.1  micron and 90% for 1 micron
 particles were demonstrated in a three-stage unit.  Induced charging or dry
 charging of particulate by charge transfer from droplets is an effective and
major collection mechanism in the fine particulate size range.  Large (100
micron) droplets give better performance characteristics than small (10 micron)
droplets.
                                       A-28

-------
SOURCE:
Shih, C. C., et al,  Emissions Assessment of Conventional  Stationary Combustion
Systems; Volume II.   Internal Combustion Sources,  February 1979.
DESCRIPTION:
Emission from gas-and oil-fueled gas turbines and reciprocating engines for
electricity generation and industrial applications are assessed in this report.
The assessment method involved a critical examination of existing emissions data,
followed by the conduct of a measurement program to fill data gaps based on
a phased sampling and analysis strategy.

MEASUREMENT:
In  the first phase of the measurement progrem, one gas-fueled gas  turbine, five
distillate-oil fueled gas turbines,  and  five diesel engines were  selected  for
testing.   Evaluation of  test results led to  the recommendation for additional
tests  to determine S03 and organic  emissions from diesel  engines  which were
subsequently conducted at three  of  the  diesel  engine  sites previously tested.
The results of the emissions assessment indicate  that internal combustion
 sources contribute  significantly to the national  emissions burden.  NO ,  hydro-
                                                                       /\
 carbon and CO  emissions  from internal  combustion  sources account for approximately
 20 percent, 9  percent,  and  1 percent of the emissions of these pollutants  from
 all stationary sources.   The source severity factor,  defined as  the ratio  of the
 calculated maximum ground level  concentration of the pollutant species to the
 level at which a potential  environmental hazard exists, was used to identify
 pollutants of environmental concern.
                                          A-29

-------
 SOURCE:
 Smith,  E.  M.,  Little, Authur  D., Sensitized Fluorescence for the Detection
 of  Polycylic Aromatic Hydrocarbons, September 1978.

 DESCRIPTION:
                                       Summary
 A fluorescent  spot test has been devised for polycyclic aromatic hydrocarbons
 (PAH) based on  the sensitization of the inherent fluorescence of such compounds.
 The basic  procedure  involves  spotting a filter paper with a small amount of
 the sample solution, adding napthalene, in solution, to the spot and visually
 observing  the  fluorescence under illumination with a simple ultraviolet light
 source.
 This project was initiated to determine whether the phenomenon of fluorescence
 could be utilized in the analysis of polynuclear aromatic hydrocarbons (PAH)
 as  a class.  A major objective was to develop a simple procedure for detection
 of  PAH at  much  lower levels than current methods based on fluorescence analysis.
 This procedure  requiring only instrumentation readily available to most
 laboratories, would provide a low cost screening technique to determine whether
 environmental assessment samples contained levels of PAH such that more detailed
 analyses should be undertaken.

                               Suggested Applications
 In  its present form the sensitized fluorescence spot test is useful for
 screening  environmental assessment samples for the presence of PAH at least as
 low as 10  ug/L  (pg/uL) in solution.  The absence of sensitized fluorescence
might well indicate that additional analyses for PAH are not necessary; on the
other hand, a positive fluorescence test might indicate that GC/MS analyses
should be  performed to determine the exact nature of the PAH detected.
                                        A-30

-------
SOURCE:
Pefley, R.  K.,  Saad,  M.  A.,  et al,  Study of Decomposed Methanol  As A Low
Emission Fuel  - Final Report, April  30, 1971.
DESCRIPTION:
Studies were conducted to evaluate blends of pure and decomposed methanol
(2\\y + CO) as fuels for reducing automotive 1C engine air pollution.  These
investigations included laboratory 1C engine tests and analysis, and preliminary
design study of possible methanol decomposition chambers with associated engine
air-fuel (A/F) ratio controls.
MEASUREMENT:
Steady-state performance and emission  tests were made on a variable compression
ratio  CFR engine operating at 900 RPM.  A  total of 191 tests were  conducted.
They included  184  tests with methanol  blends and seven comparative gasoline
fueled tests.  Engine test variables were  A/F  ratio,  percent methanol  dissocia-
tion,  compression  ratio  (CR), spark advance, and intake manifold temperature.
INSTRUMENTATION:
Instrumentation  consisted of apparatus for measuring  air  and  fuel  flow rates,
engine load,  engine  emissions  and  various  engine  temperatures.   Emission
instrumentation  included CO, C0?,  and  NO  gas  analyzers  and  a gas chromatograph
                                £•        ^
 (GC)  using  a  flame ionization detector.
                                           A-31

-------
SOURCE:
Wilson, R. R., et al, Guidelines for Particulate Sampling in Gaseous Effluents
from Industrial Processes, January 1979.

DESCRIPTION:
This guideline document lists and describes briefly many of the instruments and
techniques that are available for measuring the concentration or size distribu-
tion of particles suspended in process streams.  The standard, or well
established, methods are described as well as some experimental methods and
prototype instruments.
Descriptions of instruments and procedures for measuring mass concentration,
opacity, and particle size distribution are given.  Precedures for planning and
implementing tests for control device evaluation are also included.
                                       A-32

-------
                                     APPENDIX B-l
                       ANALYTICAL EQUIPMENT CHARACTERISTICS
   COMPOUNDS:      S02, CO, C02,  NH3,  hydrocarbons
MANUFACTURER:      MSA   Instrument Division
       MODEL:      LIRA Luft-Type Infrared Analyzer 303

     PERFORMANCE
             Reproducibility: +1% of full scale
             Noise:   <1% of f-11  scale
             Lag Time:   Response time is 5 seconds (fast response time models available)
             Zero  Drift: <1% in 24 hrs.
             Span  Drift: <1% in 24 hrs.

     OPERATION
             Ambient  Temperature  Range:  5°  to 49°C   (40° to  120°F)

      REQUIREMENTS
             Power:  110 V,  60  Hz, 60 W
             Weight:      17 Kg (37 Ib)
             Dimensions:  21  cm H, 26  cm W, 52  cm D  (8-1/4" X 10-1/2"  X 20-7/16")

      FEATURES
             Output:      Meter, recorder output 50-100 mV

      COST
             Model 303   $1,935
                                        B-l

-------
   COMPOUNDS:      CO, C02, S02, NH3, NO, N02 and hydrocarbons  by  changing  cells
MANUFACTURER:      MSA   Instrument Division
       MODEL:      LIRA Luft-Type Infrared Analyzer 202

     PERFORMANCE
            Accuracy:    +_ 1% of full  scale
            Reproducibility: +. 1% of full scale
            Linearity:  Usually within 5%, always within 10%
            Noise:  <1% of full  scale
            Lag time:    24 sec  - Response time is 5 seconds (fast  response  time
            models  available)
            Zero Drift:  
-------
                                                     Cost   Cont'd.
Model  202S, with separate sheet metal  case for 20"  cells,  general  $3,350
Model  202SX, with separate explosion proof case for 20".cell,
       explosion proof   $4,415
Model  202FR, Fast Response, general purpose $2,855
                                  B-3

-------
   COMPOUNDS:       S02,  CO and  C02
MANUFACTURER:       Leeds & Northrup
       MODEL:       7864  and 7865 Infrared Analyzers

     SAMPLING
            Maximum Temperature Input:   93°C  (200°F)

     PERFORMANCE
            Accuracy:  +_ ]% of  span  or + 0.003% of gas (whichever is greater)
            Reproducibility:  +  1.5£  of span or j^ 0*003% of gas (which ever is greater)
            Lag Time:  1  sec.
            Zero Drift:  + 0.72/24  hr & +2%/7  days
            Span Drift:  + O.H/24  hr & 0.3%/7 days

     OPERATION
            Ambient Temperature Range: 4° to  49°C ( 40° to 120°F)

     REQUIREMENTS
            Power:  108-132 V, 50/60  Hz
            Weight:      31.7  Kg (70  Ib)
            Dimensions:  37 cm W, 108 cm  H, 30 cm D (14-5/8" X 42-3/8" X 11-5/8")

     FEATURES
            Output:      0-10  mV standard, current outputs 0-16 mA, 4-20 mA, 0-20 mA,
                        0-40  mA, and 10-50 mA.
                                           B-4

-------
   COMPOUNDS:      Several, by change or use of  different  sensor
MANUFACTURER:      RFC
       MODEL:      Biosensor Vapor Detection System,  1332-16

     SAMPLING
             Method:     Continuous
             Volume:     300-600 cc/nrin
             Collection Efficiency:  Maximum possible

      PERFORMANCE
             Accuracy:   Not computed
             Reproducibility:  +  20%
             Linearity:  Not calculated
             Noise:  40%  of minimum detectable sensitivity
             Lag Time:     <5 sec for gaseous pollutants
             Retention Time:  Function of concentration, usually  <5 sec
             Fall  Time:  <10 sec for low concentration;<60 sec for  crash concentration
             Zero Drift:  Not determined
             Span Drift:  Self zeros, time not determined

      OPERATION
             Ambient Temperature  Range:  0° to 32°C (32° to 90°F)
             Temperature  Compensation:   Internal
             Relative Humidity  Range:   40-90% R.H.
             Procedure:   Self explained in  operation manual
                                              B-5

-------
                                                         Operation   Cont'd.
       Unattended Period:  -0-
       Maintenance:  Minimal with dust and participate cleaning performed
               on a weekly basis
REQUIREMENTS
       Power:  110 Vac, 60 Hz; 220 Vac optional
       Weight:  4Kg (10 Ib)
       Dimensions:  Control unit 18 cm H, 23 cm W,  18 cm D (7"  X 9"  X 7")
        (Sensing unit and handle 8 cm H,  15 cm W, 11  cm D + 9 cm probe)
         (3" X 6" X 4-1/2" + 3-1/2" probe)

FEATURES
       Output:  Meters, lights, audible alarm and recorder output
       Training:  in-house or in-field

COST
       Model 1332-16  $6,830
                                       B-6

-------
   COMPOUNDS:      S02,  NO,  CO,  C02>  and total  hydrocarbons
MANUFACTURER:      Infrared  Industries
       MODEL:      Series 700
     SAMPLING
            Method:     Continuous-pump
            Volume:     1 CFH
            Maximum Temperature Input: Below due point

     PERFORMANCE
            Accuracy: 1% of full scale
            Reproducibility:  0.5% of full scale
            Linearity:  +_1% of full scale
            Noise: < ~\% of full scale
            Fall Time: <.5 sec.
            Zero Drift:  <5%/24 hr, 
-------
REQUIREMENTS
       Power:  115 V, 60 Hz, 50 U
       Weight:  11 Kg (25 Ib)
       Dimensions:  23cm H, 43cm W, 43cm D (  9"  X 17"  X 17")

FEATURES
       Output:  0-100 mV
COST
       Series 700          $1500
                                       B-8

-------
   COMPOUNDS:      502,  NO,  co>  C02»  opacity
MANUFACTURER:      Environmental  Data Corporation
       MODEL:      Diga-Series In-Situ Emission Monitors

     SAMPLING
            Method:     Continuous -  measures inside the stack,  no sampling system required
            Maximum Temperature Input:  As required - no limit
            Collection Efficiency:  Integrated analysis across stack or duct

     PERFORMANCE
            Accuracy +_ 2%
             Reproducibility:  +_ 1%
             Noise: typically  <2%
             Lag Time: Instantaneous - no sample system
             Zero  Drift: <2% in 30 days
             Span  Drift: 
-------
   COMPOUNDS:      S02, N02, NOX, CO, CH20  Sy replacing sensor
MANUFACTURER:      Dynasciences Corporation
       MODEL:      Carbon Monoxide Monitor Model  CO-530

     SAMPLING
            Method:      Continuous-pump in optional  Sampling unit
            Volume:      0.24 to 1.0 liters/min (0.5  to 2.0 CFH)
            Maximum Temperature Output:   43°C (110°F)
            Collection Efficiency:     Not applicable

     PERFORMANCE
            Accuracy:   + 2% of full scale
            Reproducibility:  + 2%
            Linearity:< 1/2% over entire range
            Noise:< 0.1% rms of full scale
            Lag Time:   0.1  sec
            Retention  Time:  0.1  sec
            Zero Drift: 1% in 24 hr
            Span Drift: 1% in 24 hr

     OPERATION
            Ambient Temperature Range: 4° to 43°C (40° to 110°F)
            Temperature Compensation:  Compensated to +_ 10% between 4° to 43°C
                     (40°  to 110°F) for optimum accuracy, temperature control units
                     are recommended.
                                            B-10

-------
                                                             Operation Cont'd.
       Relative  Humidity  Range:  0- 99% at 37.8°C
       Unattended  Period:   1 week minimum
       Maintenance:   Weekly zero and  span checks are  recommended

REQUIREMENTS
       Power:   115/230 V, 50/60 HZ, option  12 Vdc
       Weight:  5.9 Kg (13 Ib)
       Dimensions: 21 cm H, 28  cm  W,  28 cm  D  8.25" X 11"   X 11")

FEATURES
       Output:     meter, 0-10 mV  recorder connection
       Training:   not required

COST
       Monitor Model CO-530         $2,395
       Sampling Unit Model CXS-1000  1,680
                                        B-ll

-------
   COMPOUNDS:      CO
MANUFACTURER:      Devco Engineering Inc.
       MODEL:      Series 10 APM Environmental  Co.

     SAMPLING
            Method:  Continuous (Sample is drawn through a series of filters &
                     scrubbers to remove particulates & hydrocarbons)

     PERFORMANCE
            Accuracy:  +_0.5%
            Reproducibility:  + 1% of full scale
            Linearity:   1.0% of full scale
            Noise:  +0.5%
            Lag Time:  Approximately 15 sec
            Zero Drift:  
-------
   COMPOUNDS:      CO,  C02,  CH4,  C2H6>  C2H4,  C2K2,  C3H8,  C4H0,  C6H14,  NH3,  N20, NO.
                    et al.,  dependent on filter
MANUFACTURER:      Bendix Corporation
       MODEL:      Infrared Gas Analyzer UNOR 2

     SAMPLING
            Method:     Continuous-pump or external pressure
            Volume:     0.47 liters/min
            Maximum Temperature Input:  35°C
            Collection Efficiency:  100%

     PERFORMANCE
            Accuracy:  +_ }% of full  scale
             Reproducibility:  +_ 1% of full scale
             Lag Time:  2 sec
             Zero  Drift:  +  1%  of  full scale
             Span  Drift:  +_ 1%  of  full scale

     OPERATION
             Ambient Temperature  Range:  15  to 35°C
             Temperature  Compensation:   Achieved by a  temperature dependent feed-
                    back  to amplifier
             Relative Humidity  Range:  To 98%
             Procedure:  Assembly is  gasketed and can  be air purged & interlocked.
             Maintenance:   Supply reference gas N2 (sealed or flowing)
                                            B-13

-------
REQUIREMENTS
       Power:  110 V, 60 Hz, 20 W
       Weight:  22 Kg (49 Ib)
       Dimensions:  51 cm H, 37 cm W, 15 err, D (20" X 14-1/2" X 6")

FEATURES
       Output:  Meter, recorder connection 0.1-1.0 mA (live zero output)

COST
       UNOR 2 with one range      $3,403
       UNOR 2 with dual  ranges    $3,685
                                      B-14

-------
   COMPOUNDS:       CO, CH4 and total  hydrocarbons
MANUFACTURER:       Bendix Corporation
       MODEL:       Series 8200 Environmental  Chromatograph
     SAMPLING
            Method:     Continuous (cyclic operation)-pump
            Volume:     200 cc per min.
            Collection Efficiency:  100%

     PERFORMANCE
            Accuracy:   + 1% of full scale
            Reproducibility:  + 1% of full scale
            Linearity:  + 0.5%
            Noise:      <0.01 ppm
            Lag Time:    Does  not apply.   5 min cycle  time.
            Retention Time:     Does  not  apply.  5 min cycle  time.
            Fall  Time:   Does  not apply.   5 rn'n cycle  time.
            Zero  Drift:   +  1% for  24 hr.  or  +_ 2% for  3  days
            Span  Drift:   +  1% for  24 hr  or + 2% for 3 days

      OPERATION
            Ambient Temperature Range:  5° to 40°C
             Temperature Compensation:  Limit ambient fluctuation to + 5°C.   Sample
                  loops, injection valves and columns are controlled to +_ 0.5°C.
             Relative Humidity Range:  to 95%
                                              B-15

-------
                                                            Operation Cont'd.
       Procedure:  With gas of known concentration
       Unattended Period:  7 days
       Maintenance:   Provide gases for operation.
       Periodic replacement of sample participate  filter.

REQUIREMENTS
       Power:   115 + 10 V, 50 or 60 Hz as specified 500 W, 700 VA Max.
       Weight:  approximately 115 Kg (250 Ib)
       Dimensions:  107 cm H, 56 cm W, 58 cm D (41" X 22"  X 23")

FEATURES
       Output:  Meter, recorder connections )-!DmV + 0-1 V DC
       Training:  Factory training course

COST
       Series  8200 Environmental Chromatograph      $6,900
                                       B-16

-------
                                                        Operation Cont'd.


       Procedure:   Chart  supplied with  instrument which gives a specific in-
                      strument  reading  when inserted instead of tape cassette.
       Unattended  Period:    168 hours  (1 week of 24 hour operation)
       Maintenance:   None required  except  to change cassette once per week

REQUIREMENTS
       Power:   115 V, 60  Hz, 35 W
       Weight:  10  Kg (22  Ib)
       Dimensions:  24 cm H, 41 cm  W,  25  cm D  (9-1/4"  X 16"  X  10")

FEATURES
       Recorder Output:  0-1 mA dc or 0-100 mV  dc
       Training:         None

COST
       Model 7050    $2,465
                                          B-17

-------
   COMPOUNDS:       The Model  7050 can be modified to also measure H2S,  N02,  Phosgene,
                    Chlorine, Amonia and T D I.
MANUFACTURER:       Universal  Environmental Instruments  (U.K.)  Ltd.
       MODEL:       Model  7050 CO Detector

     SAMPLING
            Method:     Continuous, with self-contained pump,  needle valve & flow meter
            Volume:     500 cc/min
            Collection Efficiency: 100%

     PERFORMANCE
            Accuracy:  +_ 10% of reading
            Reproducibility:   Better than 5%
            Linearity:  Better than 5%
            Noise:       1.055
            Lag Time:   30 seconds
            Retention Time:  30 seconds
            Fall Time:  30 seconds
            Zero Drift: Better than 1% for 24 hours
            Span Drift: Better than 1% for 24 hours
     OPERATION
            Ambient Temperature Range:  0 to 40°C
            Temperature Compensation:   None
            Relative Humidity Range:  0-100% RH
                                           B-18

-------
   COMPOUNDS:      CO
MANUFACTURER:      MSA Instrument Division
       MODEL:      Portable CO Indicator Model D

     SAMPLING
             Method:     Continuous-pump
             Volume:     Approximately 1.5 liters/min.

     PERFORMANCE
             Accuracy:  +_  1% of full scale
             Reproducibility:  +_  5%
             Lag Time:  Response  time  -50 sec  (90%  of value) with  5  ft  sample  line
             Zero  Drift:   +  1%/10 min

      REQUIREMENTS
             Power:   Battery powered - 72  volts
             Weight:   3 Kg (7-1/2 Ibs)

      COST
             Model D     $367.50
             Battery Charger  $42.00
             Alarm Unit  $210.00
                                             B-19

-------
   COMPOUNDS:      CO
MANUFACTURER:      Matheson Gas Products
       MODEL:      Carbon Monoxide Detector Models 8031, 8032, 8033

     PERFORMANCE
            Accuracy:  <+_ 1% of full  scale

     COST
            Model 8031           $1545
            Model 8032           $1545
            Model 8033           $1545
                                           B-20

-------
   COMPOUNDS:  NO
MANUFACTURER:  Enraf-Nonius
       MODEL:  TNO Ambient Monitor NO

     SAMPLING
            Method:       Continuous
            Volume:       2.4 liter/hour
            Collection Efficiency:  Better than 95%

     PERFORMANCE
            Linearity:  Better than 2% of full scale
            Lag Time:   Integrated NO concentration is measured over periods of a 1/2 hour.
            Retention Time:   Integrated NO concentration is measured over periods of a 1/2 hr.
            Fall Time:  Integrated NO concentration is measured over periods of a 1/2 hr.
            Zero Drift:  Compensation provided for
            Span Drift:   < 5% in 3 months

     OPERATION
             Ambient Temperature Range:  0° to 30°C
            Relative Humidity Range:  10 to 100%
            Procedure:  Reagent consumption is 10 liters per 3 months
            Unattended Period:  3 months
            Maintenehce:   Twice per annum

     REQUIREMENTS
             Power  115  V, 60  Hz; 220 V,  50 Hz;  150 W
             Weight:   30kg  (measuring  unit)
                                               B-21

-------
                  7 Kg (reagent storage container)
       Dimensions:  45 cm H, 52 cm W, 40 cm D (measuring unit)
                    45 cm H, 24 cm W, 40 cm D (reagent storage container)
FEATURE
       Output:   0-100 mV
                                         B-22

-------
   COMPOUNDS:       S02 and  N02  in  the  same  unit,  NOX,  CO,  and  CH20 by  replacing sensor
MANUFACTURER:       Dynasciences Corporation
       MODEL:       Multi-Pollutant Gas Analyzer NS  410

     SAMPLING
            Method:     Continuous-sampling capability within unit
            Volume:  0.24 to 1.0 liters/min (0.5 to 2.0 CFH)
     PERFORMANCE
            Accuracy:  +_ 2%
            Reproducibility:  j^ 0.5% minimum
            Linearity:  j^l/2 % over entire range
            Noise: £0.1  rms of full scale
            Lag Time:  0.2  sec
            Retention  Time:  0.2  sec
            Fall  Time:  <5  min
            Zero  Drift:  +_ 2%/24  hr
            Span  Drift:  +  2/24 hr

      OPERATION
             Ambient Temperature Range:  4.4° to 43°C (40° to 110°F)
             Temperature Compensation:   +_ 10% (40° to 110°F)
             Relative Humidity  Range:   0 -  99% at 37.8°C (100°F)
             Unattended Period:  up to 30 days
             Maintenance:   Weekly  zero and span checks are recommended.
                                              B-23

-------
REQUIREMENTS
       Power:  115/230 V, 50/60 Hz
       Weight:  27 Kg (60 Ib)
       Dimensions:  28 cm H, 49 $L W,  54 cm D (IT  X  19-1/2"  X  21-1/2")

FEATURES
       Output:  meter, two recorder outputs 0-1  Vdc
       Training:  Not required

COST
       Model NS 410          $6,450
                                      B-24

-------
   COMPOUNDS:      S02,  N02>  NOX>  CO,  C^O  by  replacing  sensor
MANUFACTURER:      Dynasciences  Corporation
       MODEL:      Nitrogen Dioxide Monitor NR-230
     SAMPLING
            Method:     Continuous-pump in optional  sampling unit
            Volume:     0.24 to 1.0 liters/min (0.5 to 2.0 CFH)
            Maximum Temperature Input:  43°C (110°F)

     PERFORMANCE
            Accuracy:  + 2% of full scale
             Reproducibility:  + 2%
             Linearity: 
-------
                                                            OPERATION Cont'd.
       Relative Humidity Range: 0 - 99% at 37.8°C (100°F)
       Unattended Period:  1 week minimum
       Maintenance:  Weekly zero and span checks are recommended

Requirements
       Power; 115/230 V, 50/60 Hz^ option 12 Vdc
       Weight:  5.9 Kg (13 Ib)
       Dimensions:  21 cm H, 28 cm W, 28 cm D (8.25" X 11" X 11")

FEATURES
       Output:     Meter, 0-10 mV recorder connection
       Training:   Not required

COST
       Monitor Model NR-230       $2,000
       Sampling Unit Model  CXS-1000  $1,680
                                       B-26

-------
   COMPOUNDS:      NO /S0«  NO  /NOa  or  N02/S02
                    /\    *•   /\
MANUFACTURER:      Dynasciences  Corporation
       MODEL:      Single Gas Analyzer  Models P100R & P100D
                   Dual  Gas Analyzer Models  P101R & P101D

     SAMPLING
            Method:     Continuous-unit includes all necessary equipment to sample,
                        condition and monitor source gases.  Sampling acquisitioning
                        is accomplished through the use of a sintered stainless
                        steel filter capable of removing 98% of the particles
                        above 0.7 microns & 100% of those above 1.8 microns.
                        Heated sample lines are available.

      PERFORMANCE
             Reproducibility:  +_ \%  minimum
             Linearity:  +_ 1/2%  within  any range
             Lag Time:    Response time  NOX -30  sec
                                       NO 2  -8  min
             Zero  Drift:   <+ 2%  / 24 hr.
             Span  Drift:   <+ 2%/ 24  hr.

      OPERATION
             Ambient Temperature Range:  4°  to  43°C (40° to 110°F)
             Temperature Compensation:   Temperature controlled at 43°C (110°F)

      REQUIREMENTS
             Power:  115 V, 50/60 Hz 1100 W maximum
             Weight:  39 Kg (85 Ib)
                                             B-27

-------
FEATURES
COST



       Model  P100R       $4,950



       Model  P100D       $6,400



       Model  P101R       $6,450



       Model  P101D       $7,950
                                     B-28

-------
   COMPOUNDS:       PM123 - N02  and  NO
                   PM 124 -N02  and  S02
MANUFACTURER:       CEA  Instruments
       MODEL:       PM 123 and PM 124 sensors

     SAMPLING
            Method:     Continuous
            Volume:     250 to  500  ml/min
            Collection Efficiency:   96%

     PERFORMANCE
            Accuracy:  2%
            Reproducibility:  1%
            Lag Time:  3 min.
            Zero Drift:  2% in one week
            Span Drift:  2% in one week

     OPERATION
            Ambient  Temperature  Range:  2° to 43°C (36° to 110°F)
            Temperature  Compensation:   None
            Relative Humidity  Range:  0-100%
            Unattended Period:         2 weeks
                                            B-29

-------
                                                      Operation Cont'd.









       Maintenance:  Reagent change every 2 weeks;  pump  tube  change every month;



                     general cleaning and calibration  every 6 months





REQUIREMENTS



       Power:  12 Vdc (120 V, 60 Hz with AC pack)



       Weight:  7 Kg (16 Ib)



       Dimensions:  46 cm W, 31  cm H, 25 cm D (18"  X 12"  X 10")





FEATURES



       Output:   0-100 mV or  0-10uA; no  built-in  recorder



       Training:   None required





COST



       PM 123 or PM 124 with DC  power pack   $2,690



       PM 123 or PM 124 with AC  power pack   $2,970
                                      B-30

-------
   COMPOUNDS:      Modification to S02 or N02
MANUFACTURER:      CEA Instruments
       MODEL:      PM 113 NO  Sensor
                            y\
     SAMPLING
            Method:     Continuous
            Volume:     250 to 500 ml/min
            Collection Efficiency:  96%

     PERFORMANCE
            Accuracy:  2%
             Reproducibility:   1%
             Lag Time:  3 min.
             Zero  Drift:  2% in one week
             Span  Drift:  2% in one week

      OPERATION
             Ambient Temperature  Range:  2° to 43°C
             Temperature  Compensation:  None
             Relative Humidity Range:   0-100%
             Procedure:  Controls are provided to initiate a purge/calibrate cycle
                         either manually or automatically.
             Unattended Period:  2 weeks
             Maintenance:  Reagent change every 2 weeks; pump  tube change every month;
                           general cleaning and calibration every 6 months.
                                              B-31

-------
REQUIREMENTS
       Power:  110 V, 60 Hz, 110 W
       Weight:  8 Kg (18 Ib)
       Dimensions:  30.5 cm W, 20.5 cm H,  25.4  cm D (12"  X  12"  X  10")

FEATURES
       Output:  0-100 mV or 0-10  A;  no built in  recorder

COST
       PM 113      $2,150
                                      B-32

-------
   COMPOUNDS:      S02, N02, NOX, CO, CH20 by replacing sensor
MANUFACTURER:      Dynasciences Corporation
       MODEL:      Oxides of Nitrogen Monitor, NS-130

     SAMPLING
             Method:     Continuous-pump in optional sampling unit
             Volume:  0.24 to 1.0 liters/min  (0.5 to 2.0 DFH)
             Maximum Temperature  Input:  43°C  (110°F)

     PERFORMANCE
             Accuracy:  +_ 2% of full  scale
             Reproducibility:  +  2%   ,   1%
             Linearity: 
-------
                                                          Operation Cont'd.
       Unattended Period:  1 week minimum
       Maintenance:  Weekly zero and span checks are recommended

REQUIREMENTS
       Power:  115/230 V, 50/60 Hz, option 12 Vdc
       Weight:  5.9 Kg (13 Ib)
       Dimensions:  21 cm H, 28 cm W, 28 cm D (8.25" X 11" X 11")

FEATURES
       Output:  Meter, 0-10 mV recorder connection
       Training:  Not required
COST
       Monitor Model NX-130     $2,000
       Sampling Unit Model CXS-1000  $1,680
                                       B-34

-------
   COMPOUNDS:      CO, C02, CH4, C2Hg,  C2H4,  C2H2>  C3H8,  C4H10j  C6H14, NH3,  NgO,  NO,
                   et al., dependent on filter
MANUFACTURER:      Bendix Corporation
       MODEL:      Infrared Gas Analyzer UNOR 2

     SAMPLING
             Method:     Continuous-pump or external pressure
             Volume:     0.47 liters/min
             Maximum Temperature  Input: 35°C
             Collection  Efficiency:     100%
     PERFORMANCE:
             Accuracy:   +_ 1% of  full  scale
             Reproducibility:  +_ 1%  of  full scale b, <1%  d
             Lag Time:   2 sec b,  11.7 sec d
             Zero Drift:  +_ 1% of full  scale
             Span Drift:  +_ 1%  of full  scale
      OPERATION
             Ambient Temperature Range:  15 to 35°C
             Temperature Compensation:    Achieved by a temperature dependent feed-
                                         back to amplifier
             Relative Humidity Range:  to 98%
             Procedure:  Assembly is gasketed and can be air purged and interlocked,
             Maintenance:  Supply reference gas N2 (sealed or flowing)
                                             B-35

-------
REQUIREMENTS
       Power:   110 V,  60 Hz,  20 W
       Weight:   22 Kg  (49 Ib)
       Dimensions:  51  cm H,  37 cm W,  15 cm D  (20"  X  14-1/2"  X  6")

FEATURES
       Output:      Meter, recorder connection  0.1-1.0 mA  (live  zero output)
                                       B- 36

-------
   COMPOUNDS:       CO,  NgO,  NO,  N02,  CH4, C2H2, S02 and most other  heteroatomic  gases
                   except H20 &  C02
MANUFACTURER:       Calibrated Instruments Inc.
       MODEL:       Infrared  Gas  Analyzer SC/LC

     SAMPLING
            Method:     Continuous-pump
            Volume:     2 liters/min

     PERFORMANCE:
            Accuracy:  + 2% of full  scale
            Reproducibility: <1% of full scale
            Lag Time:  0.1 sec
            Zero Drift:  +0.9%
            Span Drift:  1.0%

     OPERATION
            Ambient  Temperature Range:  0° to 40°C (32° to 104QF)
            Temperature  Compensation:   1% per °C if without thermostat control
            Relative Humidity Range:   to 82°C  (107.6°F)
            Maintenance:  Replace detector cell after 24 months or longer

      REQUIREMENTS
            Power: 95-130 V, 60 HZ;  190-260  V, 50 Hz; 50 VA; 100 VA with sampling pump
            Weight:  25  Kg  (56  Ib)
             Dimensions:   52 cm  H, 46 cm W, 27  cm  D  (20.5"  X  18"  X  10.5")
                                              B- 37

-------
FEATURES
       Output:      Meter,  chart recorder connection

COST
       Infrared Gas Analyzer SC/LC   $2,995
                                        B- 38

-------
   COMPOUNDS:       NOX with external converter
MANUFACTURER:       Bendix Corporation
       MODEL:       8102 NO - NOX Analyzer

     SAMPLING
            Method:       Continuous
            Volume:       20 to 30 cc/min and 150 cc/min oxygen
            Maximum Terperature Input:  150° to 200°C
            Collection Efficiency:   1000

     PERFORMANCE
            Accuracy:  +_ 1% of full  scale
            Reproducibility:  +1%
            Linearity:  j^ 0.5% full  scale
            Noise:  0.5 ppm
            Lag Time:  5 seconds
            Retention Time:  5 seconds
            Fall Time:  15 seconds
            Zero Drift:  + 2% for 24 hours
            Span Drift:  + 21 for 24 hours

     OPERATION
            Ambient Temperature Range:   5°  to  40°C
            Temperature Compensation:   Limit ambient fluctuations  to +_ 5°C.   Photo-
                                       multiolier tube  and  reaction chamber  w/ith
                                        analytical components  temperature controlled.
                                               B- 39

-------
                                                   Operation  Cont'd.
       Relative Humidity Range:   to 95%
       Procedure:   With Gas of known concentration
       Unattended  Period:  7 days
       Maintenance:   Provide oxygen for operation.   Periodic  replacement  of
                      sample particulate filter and  exhaust gas  activated
                      charcoal scrubber.
REQUIREMENTS
       Power:      115 +_ V, 50 or 60 Hz as  specified, 350 W
       Weight:     Approximately 27 Kg (60  Ib)
       Dimensions: 42 cm W, 22 cm H, 46 cm  D (8-1/2" X 16-1/2" X 18")

FEATURES
       Output:     0-10 mV and 0-1  Vdc
       Training:   Factory training course

COST
       Model 8102  (for NO only)   $4,250
         with converter option   $4,675
                                        B-40

-------
   COMPOUNDS:       Model  A100 -  N02
                   Model  A1001  - NO/N02/NOX
MANUFACTURER:       Freeman Laboratories,  Inc.
       MODEL:       N02/NO/NOX Continuous  Analyzer Models  A100 and  A1001

     SAMPLING
            Method:     Continuous
            Collection Efficiency:  99 +  %

     PERFORMANCE
            Lag Time:   6-15 sec
            Zero Drift: maximum 1%/day
            Span Drift: maximum 1%/day

     REQUIREMENTS
            Power: 115 V/60 Hz/0.25 amps or 220 V/50 Hz/0.125 amps
            Weight:     18 Kg (40  Ib)
            Dimensions: 30 cm H,  28 cm W,  37 cm D  (12" X 11" X 14-1/2")

     FEATURES
            Output:     Recorder

     COST
            Model  A100  (N02) Standard    $3,400
                   Self calibrating      $4,500
            Model  A1001  (NO/N02/NOX)  standard   $4,150
                   Self calibrating             $5,250
                                            B- 41

-------
   COMPOUNDS:
MANUFACTURER:      Bendix Corporation
       MODEL:      400 Gas Detector
     SAMPLING
            Method:     Intermittent - manual  pump
            Volume:     1-100 ml

     PERFORMANCE

     OPERATION
            Procedure:  Gas to be sampled is drawn through detector tube and length
                        of stain compared to concentration chart provided.
            Maintenance:  Tube replacement

     REQUIREMENTS

     FEATURES
            Output:     Length of stain

     COST
            Model 400     $90.00
                                            B-42

-------
   COMPOUNDS:      NO, NOX with N02 derived from the NOX -  NO difference
MANUFACTURER:      Bendix Corporation
       MODEL:      8101B

     SAMPLING
            Method:     Continuous
            Volume:     150 cc/min sample and 30 cc/min oxygen

     PERFORMANCE
            Accuracy:  +_ 0.01 ppm or +_ 2% whichever is greater on the 0 to 2.0 ppm scale
             Reproducibility:  +_ 0.01 ppm from   .005 to 2.0  ppm measured at integrator output
             Linearity:  +  0.5% full scale
             Noise:  +  0.5 % of full scale
             Lag Time:   5  sec
             Retention  time:  5 sec on NO mode
             Fall  time:  Less than  15 sec on NO  mode
             Zero  Drift: Less than  0.01 ppm in  24 hr. on the  2.0  ppm range  and
                         less than  0.05 ppm for 3 days  on  the 2.0 ppm range
             Span  Drift:  Less than  0.08 ppm in  24 hr. on the  2.0  pom range
                         and  less  than 0.08 ppm for 3 days  on the 2.0 ppm range
      OPERATION
             Ambient Temperature Range:   5° - 40°C
             Temperature compensation:   Limit ambient fluctuations to +_ 5°C.  Photo-
                                        multiplier tube and  reaction chamber  temperature
                                        control
                                             B- 43

-------
                                                   Operation Cont'd.


       Relative Humidity Range:   to 95%
       Procedure:   With gas of  known concentration
       Unattended Period:  7 days
       Maintenance:   Provide gases   for operation.   Periodic replacement
                     of sample participate filter and exhaust gas  activated
                     charcoal  scrubber.
REQUIREMENTS
       Power:  115 +_ 10 V, 50 or 60 Hz as specified, 350 W
       Weight: approximately 27  Kg  (60 Ib)
       Dimensions:  22 cm H, 42  cm  W, 43 cm D (8-1/2" X 16-1/2"  X  17")
FEATURES
       Output:     Meter, recorder  0-10 mV and 0-1  Vdc
       Training for Operation:  Factory training course
COST
       Model 8101B    $5,870
                                       B-44

-------
REQUIREMENTS

      Power:    115 ±10 V,  50 or 60 Hz  as  specified,  500 W,  700  VA max

      Weight:    Approximately 115 Kg (250  Ibs)

      Dimensions:    107 cm  H, 56 cm W,  58  cm D  (41"  x 22"  x  23")


FEATURES

      Input:  Meter, recorder connections  0-10  mV  and 0-1  V  dc

      Training:  Factory training course
COST
      Series 8200
      Environmental  Chromatograph            $6900
                                          B- 45

-------
    COMPOUND:        Hydrocarbons
MANUFACTURER:        Teledyne
       MODEL:        Series  400

     SAMPLING
            Method:       Continuous
            Volume:       100 ml  to 400  ml  per minute of sample
            Maximum Temperature Input:        177°C  (35QOF)
            Collection Efficiency:        1002

     PERFORMANCE
           Accuracy:       1% of  full scale range
            Reproducibility:       +1%
            Linearity:        +2%
            Noise:      <+}/2%  full scale
            Lag  Time:        15  seconds
            Retention  Time:        15 seconds
            Fall  Time:        l  second
            Zero  Drift:       less than  1*
            Span  Drift:       less than  1%

     OPERATION
            Ambient  Temperature Range:        4° to  38°C (40° to 100°F)
            Temperature Compensation:  Contains proportional temperature controller
            Unattended Period:     1 day
                                             B-46

-------
   COMPOUNDS:       CO, C02,  hydrocarbons,  nitrogen  and  sulfur  compounds
MANAFACTURER:       Varian Aerograph
       MODEL:       2732

     SAMPLING
            Method:     Cyclic-gas sampling valv.j
            Volume:     Variable
            Maximum Temperature Input:  225°C

     PERFORMANCE
            Accuracy       +}%
             Reproducibility:    +1%
             Noise:    < 5 microvolts
             Retention  Time:    Variable
             Zero  Drift:    <  10 microvolts/month

     OPERATION
             Maintenance:     Monthly-minimal

     REQUIREMENTS
             Power:  115 V,  20 amp
             Weight:   341  Kg  (155  Ib)
             Dimensions: 51 cm  H,  49 cm W,  55 cm  D (20 1/4"  x 19 3/8" x 21  3/4")

     FEATURES
             Output:   Recorder
             Training:   Optional  at extra cost

     COST
             Model 2732   $5195 to 7,395
                                               B- 47

-------
REQUIREMENTS
       Power:     115 V, 60 Hz, 200 watts (Model  403), 350 watts (Model  404)
       Weight:      Approximately 18 Kg (40 Ib)
       Dimensions:  Model  403:   46 cm H x 56 cm W x 46 cm D (18" x 22"  x 18")
                    Model  404:   51 cm H x 56 cm W x 46 cm D (20" x 22"  x 18")

FEATURES
       Output:    Meter, provision for 0*5 mV dc recorder
       Training:     Available from-manufacturer

COST
       Model 403 (for sampling systems with dew points below 52°C [125°F])   $2750
       Model 404 (for sampling systems with dew points up to 66°C [175°F])   $7500
                                      B-48

-------
   COMPOUNDS:       Total  hydrocarbons  based  on  methane equivalent
MANAFACTURER:       Wemco  Instrumentation  Company
       MODEL:       Automatic Gas Detector and Alarm System  Model MUC  12CS

     SAMPLING
            Method:  Continuous-positive vaccum  pressure
            Volume:  Approximately 70 SCFH air
            Maximum Temperature Input:   49°C (120°F)
            Collection Efficiency:    100% at sample input point at time of reading

     PERFORMANCE
           Accuracy:  +3% of full scale
            Reproducibility:   +3% of full scale
            Linearity:   +3% of full scale
            Noise:  Negligible
            Lag Time:  10 seconds through one 200 foot 1/4" tube
            Retention Time:  3 seconds through one  200 foot 1/4" tube
            Fall Time:  2 seconds
            Zero Drift:   Negligible
            Span Drift:   Negligible

     OPERATION
            Ambient  Temperature  Range:    18° to 35°C (65°  to 95°F)
             Relative Humidity  Range:   0 to 100%
             Unattended Period:   unknown
             Maintenance:  Approximately 2 hours per year  plus cylinder  changing
                                              B-49

-------
REQUIREMENTS



       Power:  120 V.  60 Hz  or 240  V,  50  Hz



       Weight:   Approximately 77 Kg  (170 Ibs)



       Dimensions:    155 cm H, 57 cm W, 65  cm  D  (61" x  22 1/2" x 25 1/2")





FEATURE



       Output:    0-lma  dc meter relay readout



       Training:  Available at nominal charges





COST



       Model MUC 12CS           $3,950



       (explosion proof model)  $4,975
                                        B-50

-------
   COMPOUNDS:        NO,  CO,  C02  and Hydrocarbons
MANUFACTURER:        Horiba Instruments  Incorporated
       MODEL:        Modular Non-Dispersive  Infrared Gas Analyzer Model AIA-21

     SAMPLING
            Method:    Continuous
            Volume:    2 to 20 CFH
            Maximum Temperature Input:     40°C (105°F)

     PERFORMANCE
            Accuracy:    ll%
            Reproducibility:    +1%
            Linearity:     5% of full scale at mid-scale (linearizer available)
            Noise:    < 1%
            Lag Time:  Response time -0.5 seconds
            Zero Drift:   1% of full scale/8 hours
            Span Drift:   1% of full scale/8 hours

     OPERATION
            Ambient Temperature  Range:   0° to 40°C  (32°  to  105°F)
            Temperature  Compensation:    Thermostatically  controlled
            Relative Humidity Range:     0-90% RH at 20°C  (68°F)

      REQUIREMENTS
            Power:    115 V,  60  Hz or 230 V, 50 Hz, 250 U
            Weight:  Analyzer Section  - 10 Kg (21 Ibs)
                      Amplifier  Section - 7 Kg (16 Ibs)
             Dimensions:    Analyzer Section -  46  cm H,  10  cm W,  16 cm D
                           (200  mm cell) (18"  x 4" x  6-1/10")
                                              B- 51

-------
                       Amplifer  Section  -  15 cm H, 24 cm W, 46 cm D
                       (6"  x  9-3/8"  x  18")


FEATURES

       Output:    0-10  V and 0-100 mV dc  for recorder and 0-1 V and 0-5 V input

                 to  Data Acquisition System
                                      B-52

-------
   COMPOUNDS:        Total  hydrocarbons, CHq and CO



MANUFACTURER:        Hewlett-Packard  Company



       MODEL:        5781  A Environmental  Analyzer





   REQUIREMENTS



            Weight:    43 Kg (95 Ibs)



            Dimensions:     35 cm H,  68 cm W,  48  cm D (13 1/2"  x  27"  x  19");



                           including detector:   47 cm H (18 1/2")





   COST



            Model 5781A                          $3,985



            opt. 34 (Total hydrocarbon analysis  capability)    $1,140
                                              B-53

-------
   COMPOUNDS:        S02,  CO,  0)3,  NH3,  hydrocarbons
MANUFACTURER:        MSA  Instrument
       MODEL:        303

   PERFORMANCE
            Reproducibility:   ^1% of full  scale
            Noise:               < 1% of full  scale
            Lag Time:          Response time  is 5 seconds  (fast response time models
                              available)
            Zero Drift:          < 1% in 24  hours
            Span Drift:          <1% in 24  hours

   OPERATION
            Ambient Temperature  Range:    5°  to 49°C   (40° to 120°F)

   REQUIREMENTS
            Power:     110 V,  0 Hz,  60 W
            Weight:    17  Kg  (37  Ibs)
            Dimensions:   21  cm H,  26 cm W,  52 cm 0(8-1/4" x 10 1/2" x 20-7/16")

   FEATURES
            Output:    Meter,  recorder output  50-100 mV

   COST
            Model 303                           $1,935
                                            B-54

-------
   COMPOUNDS:        Total  hydrocarbons only
MANAFACTURER:        MSA  Instrument  Division
       MODEL:        Total  hydrocarbon Analyzer MSA 2

     SAMPLING
            Method:    Continuous

   PERFORMANCE
            Accuracy:  1%
            Linearity: Linear over operative ranges
            Noise:    
-------
   COMPOUNDS:        Methane,  total  hydrocarbons  and  total  hydrocarbons  less
                    methane only
MANUFACTURER:        MSA Instrument  Division
       MODEL:        11-2

     SAMPLING
            Method:   Continuous

   PERFORMANCE
            Accuracy:     1% of  full  scale
            Reproducibility:     1%  of full  scale
            Linearity:           Linear over range
            Noise:    <  +1% of full  scale
            Lag Time:  Response  time  is 15 seconds,  100% of reading
            Zero Drift:   
-------
                             APPENDIX C-l
                  LIST OF ENGINE MANUFACTURERS/ENGINES
The following list includes  the  names  and  addresses of the small engine

manufacturers contacted for  information.   Information sought  included

size (hp), type (# strokes), application  (lawn,  etc.), emission  data,

fuel-air mixture, life expectancy and  duty cycle.   Responses  were

generally poor.


The list also contains the size range  (hp) of the engines produced by

each manufacturer.
NAME

1.  Briggs & Stratton
    Corporation

2.  Chrysler Corp.
3.  Clinton Engines
    Corporation

4.  Homelite  (Div. of
    Textron

5.  Jacobsen  Mfg.  Co.
 6.   Outboard  Marine
     Corp.  (Lawnboy)
 7.   McCulloch  Corp.
 8.   0 & R Engines,  Inc.
 9.   Tecumseh Prod.  Co.
ADDRESS

3300 N. 124th Street
Wauwatosa, WI  53201

Marine & Ind. Products Plant
Marysville, MI

Maguoketa, 10  52060
70 Riverdale Avenue
Port Chester, NY  10573

1721 Packard Avenue
Racine, WI  53403

Gale Products Division
100 Sea Horse Drive
Waukegan,  IL  60085

6101 Century Boulevard
Los Angeles, CA  90045

3340 Emery Street
Los Angeles, CA  90043

Ottawa &  Patterson  Streets
 1-1/8--3-9/16 bore size single cylinder engines
 2500 to 7200 rpm speed range
 300 to 500 MM $ industry
 12.5 + MM engines/year
 Average engine use 50 hrs./year
                                         C-l
ENGINE SIZES(HP)

   2-16 hp


   3-l/4-8hp


   4-7hp


   2- 4hp


      3hp


      lOhp



      20hp


   1-2  l/4hp


   2-1/2—16hp

-------
                                TECHNICAL REPORT DATA
                         (Please read Instructions on the reverse before completinx)
 \ 3EPORT NO.
 EPA-600/7-80-088
4. TITLE AND SUBTITLE
 A Research Plan to Study Emissions from Small
 Internal Combustion Engines
            6. PERFORMING ORGANIZATION CODE
                                                      3. RECIPIENT'S ACCESSION NO.
            5. REPORT DATE
             April 1980
7 AUTHOH(S)
James W. Murrell
                                                      8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Systems Research and Development Corporation
P.O. Box 12221
Research Triangle Park, North Carolina  27709
                                                      10. PROGRAM ELEMENT NO.
            INE624A
            11. CONTRACT/GRANT NO.

            68-02-3113
 12. SPONSORING AGENCY NAME AND ADDRESS
 EPA, Office of Research and Development
 Industrial Environmental Research Laboratory
 Research Triangle Park, NC  27711
            13. TYPE OF FlEPOHT AND PERIOD COVERED
            Final; 9/78-9/79
            14. SPONSORING AGENCY CODE
              EPA/600/13
15. SUPPLEMENTARY NOTESIERL-RTP project officer is John H,
541-2476.
             Wasser, Mail Drop 65, 919/
 16. ABSTRACT
          The report examines some of the requirements for investigating the envi-
 ronmental status of small internal combustion (1C) engines. These engines range in
 size from 1. 5 to 15 hp and power a variety of equipment operated by homeowners
 and industry.  With EPA's general growing concern of identifying sources of poten-
 tially carcinogenic emissions, a possibility  exists that these small 1C engines are a
 problem source. Research to characterize emissions from 1C engines has largely
 been limited to critical pollutants, even though the small 1C engine is an incomplete
 combustor. It follows that some carcinogens and other hazardous compounds are
 probable.  The basic requirements addressed in the report include analytical equip-
 ment,  experimental systems design, and statistical experimental design.
                             KEY WORDS AND DOCUMENT ANALYSIS
                DESCRIPTORS
b.IDENTIFIERS/OPEN ENDED TERMS
c. COSATl Field/Group
Pollution
Internal Combustion Engines
Analyzing
Systems Engineering
Statistical Analysis
Carcinogens
 Pollution Control
 Stationary Sources
 Analytical Equipment
 13B
 21K
 14 B
 05A
 12A
 06E
13. DISTRIBUTION STATEMENT
 Release to Public
                                          19. SECURITY CLASS (This Report)
                                          Unclassified
                         21. NO. OF PAGES
                                128
20. SECURITY CLASS (Thispage)
Unclassified
                         22. PRICE
EPA Form 2220-1 (9-73)
                                       C-2

-------