EPA-600/2-77-187a
September 1977
Environmental Protection Technology Series
         EMISSION TESTING AND EVALUATION
                       OF FORD/KOPPERS COKE
                   PUSHING CONTROL SYSTEM;
                     VOLUME I.  FINAL REPORT
                           Industrial Environmental Research Laboratory
                                Office of Research and Development
                               U.S. Environmental Protection Agency
                           Research Triangle Park, North Carolina 27711

-------
                      RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development, U.S. Environmental Protection
Agency, have been grouped into five series. These five broad categories were established to
facilitate further development and application of environmental technology. Elimination of
traditional grouping was consciously planned to foster technology transfer and a maximum
interface in related fields. The five^series are:

    1.  Environmental Health Effects Research
    2.  Environmental Protection Technology
    3.  Ecological Research
    4.  Environmental Monitoring
    5.  Socioeconomic Environmental Studies

This report has been assigned to the ENVIRONMENTAL PROTECTION TECHNOLOGY
series. This series describes research performed to develop and demonstrate instrumenta-
tion, equipment, and methodology to repair or prevent environmental degradation from point
and non-point  sources of pollution. This work provides the new or improved  technology
required for the control and treatment of pollution sources to meet environmental quality
standards.
                            EPA REVIEW NOTICE

This report has been reviewed by the U.S. Environmental Protection Agency, and approved
for publication. Approval does not signify that the contents necessarily reflect the views and
policy of the Agency, 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 Information Service,
Springfield, Virginia 22161.

-------
                                       EPA-600/2-77-187a
                                          September 1977
EMISSION TESTING AND EVALUATION
        OF  FORD/KOPPERS COKE
      PUSHING CONTROL SYSTEM;
       VOLUME I.  FINAL REPORT
                        by
                   Fred Cooper, Thomas Loch,
                 John Mutchler, and Janet Vecchio
                (Clayton Environmental Consultants, Inc.)

                    Ford Motor Company
                    The American Road
                   Dearborn, Michigan 48121
                   Contract No. 68-02-0630
                  Program Element No. 1A6604
                 EPA Project Officer Norman Plaks

              Industrial Environmental Research Laboratory
                Office of Energy, Minerals, and Industry
                 Research Triangle Park, N.C. 27711
                      Prepared for


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

-------
                        ACKNOWLEDGEMENTS


The principal investigators in this project wish to acknowledge the
help and cooperation of several individuals and agencies whose assis-
tance helped assure the success of this overall program.

Messrs. Norman Plaks and Douglas Kemnitz, Project Officers in the U.S.
Environmental Protection Agency, provided continuing guidance in the
planning and execution of this study over the three-year duration of
the project.

Mr. Joe Robbins of Ford Motor Company and supporting staff at Ford
assisted immensely in the draft reviews and editorial process of
preparing this report for final publication; their cooperation is
gratefully acknowledged.

Finally, the authors wish to thank their colleagues and supporting
staff at Clayton Environmental Consultants, including the analytical
laboratory personnel, field sampling crews, statisticians, secretaries,
and all other personnel who helped prepare and assemble this final
work product.

-------
               TABLE   OF   CONTENTS

                                                             Page

GLOSSARY OF TERMS 	    V

LIST OF FIGURES	   vii

LIST OF TABLES	   viii

LIST OF APPENDICES	."....    Ix


VOLUME I

1.0  SUMMARY AND CONCLUSIONS	     1

     1.1  Overall Evaluation  of the System  	     1
     1.2  Emission Factors	     4


2.0  INTRODUCTION 	     6

     2.1  Background and Purpose  	     6
     2.2  Project Objectives  	     7

          2.2.1  Documentation of Historical Performance  .     8
          2.2.2  Documentation of Historical Economics  ...     8
          2.2.3  Determination of System Operability  	     9
          2.2.4  Measurement  of Scrubber Performance  	     9
          2.2.5  Evaluation of the System and  Its
                     Applicability 	     9


3.0  COKE PRODUCTION IN THE IRON AND STEEL  INDUSTRY AND
         AT THE FORD MOTOR COMPANY 	    10

     3.1  Description of the  Pushing Process  	    10

          3.1.1  General Description  	    10
          3.1.2  Pushing Process at Ford Motor  Company  ...    11

     3.2  Pushing Emissions 	    12


4.0  DESCRIPTION OF THE FORD/KOPPERS COKE-PUSHING  EMISSION
         CONTROL SYSTEM  	    15

     4.1  System Design	    15
     4.2  System Modifications and Observed Operation ....    16
     4.3  Operating Parameters	    19

-------
                                                            Page

5.0  FIELD INVESTIGATION 	   21

     5.1  Methodology of Data Acquisition 	   22

          5.1.1  Documentation of Historical Performance  .   22
          5.1.2  Documentation of Historical Economics ...   24
          5.1.3  Determination of System Operability 	   25
          5.1.4  Measurement of Scrubber Performance 	   26
          5.1.5  System Evaluation and Applicability 	   27

     5.2  Emissions Measurement 	   28

          5.2.1  EPA Method 5 Sampling 	   30

                 5.2.1.1  Particulate 	   31

          5.2.2  EPA Method 8 Sampling 	   31

                 5.2.2.1  Sulfur Dioxide 	   32
                 5.2.2.2  Sulfur Trioxide 	   32

          5.2.3  Sodium Hydroxide Sampling 	   32

                 5.2.3.1  Chlorine 	   33
                 5.2.3.2  Cyanide 	   33
                 5.2.3.3  Total Phenolics 	   33
                 5.2.3.4  Total Sulfur Oxides 	   34

          5.2.4  EPA Method 7 Sampling 	   34

                 5.2.4.1  Nitrogen Oxides 	   34

          5.2.5  Grab Sampling	   34

                 5.2.5.1  Acetylene 	   35
                 5.2.5.2  Carbon Monoxide 	   35
                 5.2.5.3  Ethylene and Homologues 	   35
                 5.2.5.4  Methane and Homologues 	   35
                 5.2.5.5  Total Light Hydrocarbons 	   36

          5.2.6  Activated Charcoal Sampling 	   36

                 5.2.6.1  Benzene and Homologues	   36
                 5.2.6.2  Carbon Disulfide 	   37
                 5.2.6.3  Pyridine 	   37

          5.2.7  EPA Method 11 Sampling 	   37

                 5.2.7.1  Hydrogen Sulfide 	   38

          5.2.8  Sulfuric Acid Sampling 	   38

                 5.2.8.1  Ammonia 	   38
                               ii

-------
                                                            Page

          5.2.9   Ethanol Sampling 	   39

                  5.2.9.1   Acrolein 	   39

          5.2.10  Cyclohexane Sampling 	   39

                  5.2.10.1  Acridine 	   40
                  5.2.10.2  Anthracene 	   40
                  5.2.10.3  Benzo(a+e)pyrene 	   40
                  5.2.10.4  Carbazole 	   41
                  5.2.10.5  Chrysene 	   41
                  5.2.10.6  Fluoranthene 	   41
                  5.2.10.7  Fluorene 	   41
                  5.2.10.8  9-Fluorenone 	   42
                  5.2.10.9  Naphthalene 	   42
                  5.2.10.10 Fhenanthrene 	   42
                  5.2.10.11 Pyrene	   42

          5.2.11  Integrated Sampling 	   42

                  5.2.11.1  Total Reduced Sulfur 	   43

          5.2.12  Andersen Impactor Sampling 	   43

                  5.2.12.1  Particle Sizing 	   43

          5.2.13  Fugitive Emissions Measurement
                     ^ (Method 1) 	   44

                  5.2.13.1  High-Volume Sampling and Ana-
                                lytical Considerations ...   45
                  5.2.13.2  Theoretical Considerations ...   46
                  5.2.13.3  Qualifications 	   51


6.0  PRESENTATION AND ANALYSIS OF RESULTS 	   52

     6.1  Organization and Compilation 	   52

          6.1.1   Historical Data 	   52
          6.1.2   Process and Engineering Data	   54
          6.1.3   Scrubber Emissions Data 	   55
          6.1.4   Fugitive Emissions Data 	   55

     6.2  Emission and Performance Models 	   55

          6.2.1  Push Emissions 	   55
          6.2.2  Scrubber Performance 	   58
          6.2.3  System Performance	   60

     6.3  Evaluation of Hood Capture Efficiency 	   61
     6.4  Emission and Process/Control System Correla-
              tions	   64
                               iii

-------
7.0  SUMMARY OF FINDINGS 	    69

     7.1  Overall System Performance  	    70
     7.2  Hood Capture Efficiency  	    71

          7.2.1  Particulate Matter 	    72
          7.2.2  Sulfur Oxides 	    73
          7.2.3  Carbon Monoxide 	    74
          7.2.4  Oxides of Nitrogen 	    74
          7.2.5  Benzene and Homologues 	    74
          7.2.6  Total Light Hydrocarbons  	    75
          7.2.7  Methane and Homologues 	    75
          7.2.8  Ethylene and Homologues 	    76

     7.3  Scrubber and Total System Efficiencies 	    76

          7.3.1  Particulate Matter 	    77
          7.3.2  Sulfur Oxides 	.	    78
          7.3.3  Carbon Monoxide 	    79
          7.3.4  Oxides of Nitrogen 	    80
          7.3.5  Benzene and Homologues 	    80
          7.3.6  Total Light Hydrocarbons	    80
          7.3.7  Methane and Homologues 	    81
          7.3.8  Ethylene and Homologues 	    81
          7.3.9  Summary of All Contaminants 	    81
8.0  DISCUSSION OF RESULTS AND CONCLUSIONS 	   83

     8.1  Historical Performance Assessment 	   83
     8.2  Assessment of the Abatement Efficiency 	   85

          8.2.1  Hood Performance 	   85
          8.2.2  Scrubber Performance 	   86

     8.3  Applicability to Other Installations 	   90
     8.4  Limitations of the Study 	   91
9.0  RECOMMENDATIONS	    93

     9.1  Hood-Quench Car Clearances 	    93
     9.2  Redesign of Hood Structure 	    93
     9.3  Scrubber Operation 	    94
     9.4  Process and System Modifications 	    94
REFERENCE S	   96
                               iv

-------
                        GLOSSARY OF TERMS
 1.  Coke-Pushing Emissions

     An intermittent source emission lasting about 40 to 60 sec-
     onds, occurring on an irregular cycle with an average inter-
     val of 15 to 16 minutes.

 2.  Emission Factor
     ______________   |

     A ratio of rate of emission of some chemical species to the
     production rate associated with the process causing the emis-
     sion.

 3.  Coke Side of Battery

     That side of a coke oven battery from which the ovens are
     emptied of coke.

 4.  Push Side of Battery

     That side of a coke oven battery into which the pusher ram
     is inserted to empty the oven of coke.

 5.  Gross Coking Time

     The elapsed time in minutes between two successive pushes
     of the same oven.

 6.  Net Coking Time

     The elapsed time in minutes between the charging of a coke
     oven with coal and the pushing of that same oven.

 ?•  Door Leakage

     Any visible emissions observed emanating from coke-side oven
     doors, push-side oven doors,.or push-side chuck doors.

 8.  Quench Car Movement Emissions

     An intermittent source emission lasting about 15-40 seconds,
   -  from the end of a coke oven push to commencement of coke
     quenching.

 9.  Degree of Greenness of a Coke Oven Fash

     The subjective, visual estimate of the particulate matter
     released during a single coke oven push.

10.  Clean Fush

     A source emission on the coke side which is resultant from
     the pushing of an oven whose  charge has been carburized

-------
                        GLOSSARY OF TERMS
                           (Continued)


     completely enough so as to produce only slight quantities
     of visible particulate throughout the push duration.

11.   Green Push

     A source emission on the coke side which is resultant from
     the pushing of an oven whose charge has been carburized in-
     completely enough (throughout its entire mass) ao as to pro-
     duce copious quantities of visible particulate and/or flame
     throughout the push  duration.

12.   Fugitive Particulate Emissions

     Particulate emissions which escape capture from the
     hood and pass unrestrained into the atmosphere.

13.   Total Pushing Emissions

     The sum of the inlet of the control system particulate emis-
     sion captured and entering the control system and the fugi-
     tive particulate  emission.

14.   Filterable Particulate

     Material captured at a specified temperature,  pressure,  and
     chemical activity, on or before the first filter in a particu-
     late sampling train.

15.   Total Particulate

     Material captured at a specified temperature,  pressure,  and
     chemical activity, including condensibles,  in an entire par-
     ticulate sampling train.
                               vi

-------
                         LIST OF FIGURES

5.1.1-1   Specimen of Record for Operating Checks by Battery
              Foreman Once Each Shift for Control System

5.1.1-2   Specimen of Record for A-Door Machine Downtime

5.1.1-3   Specimen of Record for Weekly Operating Checks for
              Control System

5.1.1-4   Specimen of Data Extracted from Maintenance Department
              Logbook

5.1.1-5   Specimen of Data Extracted from Operations Department
              Logbook

5.1.3-1   Specimen of Record for Hood Performance

5.1.3-2   Schematic Diagram of Engineer-Observer During a Push

5.1.4     Specimen of Record for Scrubber Operating Parameters

5.1.5-1   Specimen of Record for Push Times and Condition

5.1.5-2   Specimen of Record for Coal Charge Data

5.1.5-3   Specimen of Record for Coke Oven Flue Temperatures

5.2       Scrubber Inlet and Outlet Sampling Locations Pushing
              Emissions Control System

5.2.13.1  Schematic Diagram of High-Volume Air Sampling Location

6.1.1     Maintenance Costs of the Koppers System

6.2.1     Schematic Diagram of Mass Balances

6.3-1     Fugitive Dust Concentration Measured Without Hood

6.3-2     Fugitive Dust Concentration Measured With Hood and
              Fan Off

6.3-3     Fugitive Dust Concentration Measured With Hood and
              Fan On

6.3-4     Hood Capture Efficiency as a Function of Average Greenness

7.3.1-1   Particle Size Distribution Scrubber Inlet

7.3.1-2   Particle Size Distribution Scrubber Outlet

8.2.1-1   A Clean Push With Control System Operating

8.2.1-2   A Moderately Green Push With Control System Operating

8.2.1-3  | A Green Push With Control System Operating
                               vii

-------
1.2

3.2-1


3.2-2

6.1.1-1

6.1.1-2

6.1.1-3
6.1.3-1
  through
  6.1.3-92

6.1.4-1
  through
  6.1.4-5

7.0-1
  through
  7.0-14

7.3.1-1
  through
  7.3.1-14

7.3.1-15
7.3.1-16
  through
  7.3.1-17

7.3.9-1

7.3.9-2


8.2.1
             LIST OF TABLES

Charge Rate and Emission Factor Conversion Equations

Emission Factors for Uncontrolled Metallurgical Coke
    Manufacture

Summary of Contaminants Expected in Pushing Emissions

Summary of Average Monthly Utilities

Summary of Average Monthly Scrubber Parameters

Summary of Required Maintenance Documented in Maintenance
    Logbook

Emission Test Summary



Summary of Fugitive Emissions



Summary of Emissions and Control Efficiencies



Particle Size Distribution
Concentration of Particulate Matter Calculated from
    Particle Sizing Samples

Composition of Particulate Matter
Summary of Total Uncontrolled Emissions

Summary of Measured Emissions for Contaminants Present
    Below Limits of Detectability

Photographic Documentation
                              viii

-------
VOLUME II                  APPENDICES

     A.     Summary of Maintenance and Operations Department
            Logbooks Pertaining to the Ford/Koppers Control
            System

     B.     Summary of EPA Methods

     C.     Source Sampling Field Data Sheets

     D.     Calibrations

     E.     Sample Calculations

     F.     Project Participants

     G.     Daily Log of Fume Hood Performance

     H.     Daily Log of Scrubber System Performance

     I.     Summaries of Process and Control System Data
                               ix

-------
                  1.0  SUMMARY AND CONCLUSIONS
In 1970, Ford Motor Company contracted with the Koppers Company
to design and construct an emission control system to abate coke-
pushing emissions at its Dearborn, Michigan steel-making opera-
tions.  In 1973, the United States Environmental Protection Agency
(EPA) contracted with Ford Motor Company to evaluate and document
the performance, reliability, and operating costs of the installed
Koppers system as it represents an important contribution to the
developing state-of-the-art in coke-pushing emission control tech-
nology.

Thereafter, Ford Motor Company engaged Clayton Environmental Con-
sultants, Inc., as subcontractor, to perform required measurements,
observations, and analyses necessary to provide a quantitative and
objective evaluation of the Koppers system with regard to perform-
ance efficiencies, reliability, operability, and maintainability.
This report summarizes that evaluation, and provides a record of
system performance based upon these measurements and the long-term
observation of day-to-day operating characteristics.  In addition,
the Koppers Company was subcontracted to write a system design
manual entitled "Coke Oven Smokeless Pushing System Design Manual,"
September, 1974.  This manual is available as a published document
from the Office of Research and Development, U.S. Environmental
Protection Agency.  The Environmental Protection Technology Series
number is EPA-650/2-74-076.


1.1  Overall Evaluation of the System

     The results of this study, obtained by sampling and analysis,
     still photography, and extensive visual observation, indicate
     that a 41- to 66-percent reduction in filterable particulate
     emissions and a 20- to 33-percent reduction in condensible
     emissions can be realized from coke pushing at the Ford coke
     battery when using the Koppers system to control a variety of
     pushes ranging from "clean" to "green."  These reported
     ranges in efficiency are based upon average degree-of-green-
     ness ratings obtained during multi-push particulate tests
     (i.e., 16 or 24 pushes per test).  Because average degree-of-
     greenness for the multi-push particulate sampling tests in-
     cluded in the study  was  never less than 1.4 nor greater than
     2.4 (on a scale of one to three for "clean" and "green" pushes,
     respectively), the reported ranges may not reflect system per-
     formance for "very clean" or "very green" pushes.  While multi-
     push particulate tests cannot, therefore, indicate performance
     for a single push, indirect estimates based on statistical
     analyses of available data, indicate that hood capture effi-
     ciencies may range from 30 percent (green push) to 92 percent
     (clean push) using visual estimating techniques and from 22
     percent (green push) to 54 percent (clean push) using hi-
     volume air sampling techniques to estimate the extent of
     fugitive emissions.  Of the push particulate emissions

-------
                          -  2  -


captured in the control system hood,  99.3 percent,  on  the
average, are  removed by the scrubber.  Because  the  reported
range in control system performance is dependent on  push
greenness, overall performance can be extrapolated  to  another
similarly configured coke battery only to the extent that  the
sequence and  nature of push-plumes sampled in this  program
are typical of another coke battery's operation.

The estimated range in hood capture efficiency  at a  given
degree-of-greenness derived from the  two techniques  (22 to
30 percent for green pushes and 54 to 92 percent for clean
pushes) represents only an  estimate of actual fugitive emis-
sions.  These estimates of  fugitive emissions form  a multi-
plicative factor for converting measured emissions  in  the
scrubber inlet duct where sampling was conducted to  total
pushing emissions.  This  factor is therefore contained im-
plicitly in all emission  factors.  Thus, the evaluation of
hood capture  efficiency,  and  therefore, of total system
abatement efficiency, is  completely dependent upon  the
accuracy and  precision of the methods for estimating fugi-
tive emissions.  Moreover,  all expressions of emission fac-
tors must be  qualified by the limitations of the sampling
technique and used cautiously in attempting to  predict emis-
sions expected from similar batteries.

The measured  collection efficiency of the scrubber  for fil-
terable particulate from  the  push emissions actually captured
by the movable hood ranges  from 99.1  to 99.6, and averages
99.3 percent.  Scrubber collection efficiency in abating con-
densibles ranges from zero  to 90.5 percent, and averages 47.7
percent.  The scrubber outlet plume opacity ranges  from nil
for "clean" or "moderately  green" pushes to about 40 percent
or greater for "green" pushes.  In light of the relatively
high scrubber efficiencies, the exit  gas opacity is directly
attributable  to the very high inlet particulate loadings that
are characteristic of green pushes.

The ability of the system to  function well in an environment
realistically typical of coke batteries is an important issue,
and this evaluation provides  a representative context in which
to appraise the system.   The  inevitability of green pushes in
any coke battery requires that a coke-pushing control system
be able to maintain collectability and extractability to a
degree that will conform with reasonable regulations.  The
information and data gathered in this study indicate that  the
system does not capture  emissions in an efficient manner for
moderately green or green pushes.  However, of the emissions
captured during a push,  the scrubbing efficiency for filter-
able particulate is excellent.

In addition to particulate matter,  emission source testing
was conducted for 31 different chemical compounds.   Of these,
most were found below the limits of detectability for the sam-
pling and analytical procedures used on the airstream entering

-------
                           3
the scrubber, where concentrations would be expected to be
the greatest.  Of those materials detected, the following
contaminants have been considered in evaluating the ability
of the control system to both capture and remove them:  sul-
fur dioxide, sulfur trioxide, total sulfur oxides, carbon
monoxide, nitrogen oxides, benzene and homologues, total
light hydrocarbons, methane and homologues-, and ethylene and
homologues.  For many of these materials, total system abate-
ment efficiency varies widely.  This is largely attributable
to emission measurements performed in exhaust airstreams con-
taining very low concentrations of these contaminants.  Gener-
ally, average overall system efficiencies, including estimated
hood capture, range from nil to 60 percent for these materials,
Replicate tests were conducted for each contaminant, some of
which were multi-push samples including a variety of push
types ranging from "clean" to "green."

Maintenance records and on-site system observations indicate
that the emission control system is prone to 20-percent down-
time which results especially during periods when a chronic
failure occurs that is difficult to diagnose.  Based upon sys-
tem observation, the redesign of several system components
would likely reduce downtime significantly.

The costs associated with maintenance, including manpower and
replacement parts, show no obvious trend to increase as the
system ages.  Examination of maintenance and cost accounting
records that document system maintenance shows that average
maintenance costs amount to about $0.15 per ton of coke pro-
duced on the 58-oven battery.

The utility requirements can vary widely, depending upon sys-
tem operation.  By design, the system operates in an idle mode
between pushes, depending upon operator discretion.  Once the
coke guide operator establishes alignment with the next oven
.to be pushed and activates the system, fully operational-mode
status is reached, requiring about 2000 HP of fan power.  De-
pending upon scheduling and the number of pusher machines and
quench cars available to a particular battery in a multi-
battery facility, the utility costs may vary substantially.
Based upon prolonged observation at the Ford Motor coking
facility, electrical energy requirements average about 30
KW-HR per ton of coke produced on the 58-oven battery.

In summary, the system, in its current configuration, abated
pushing emissions during the test program with an efficiency
estimated to range from 41 to 66 percent for filterable par-
ticulate, the prime contaminant.  Nevertheless, the control
system eliminated only about one-half of normal-operation
emissions produced by coke pushing at the time of the study.

Although the system was not operated in accordance with its
original design during the testing and evaluation program, we
view these results as both realistic and representative.

-------
                              - 4 -
     Undoubtedly,  the harsh environment and occurrence of green
     pushes,  somewhat characteristic of even a new coke battery,
     are responsible for many of the system modifications that
     have resulted in performance somewhat less than that asserted
     in the Koppers Design Manual.   In our opinion, the mobile
     hood,  as designed,  would not efficiently capture pushing
     emissions resultant from a moderately green or green push.
     We believe that capture efficiency can be improved measurably,
     however, if the hood dimensions and exhaust volume rate are
     increased substantially.
1.2  Emission Factors

     All emission factor estimates  contained herein that charac-
     terize total push emissions are  necessarily based upon and
     limited by the technique used  to estimate fugitive emissions.
     The degree to which the fugitive emission estimate is biased
     directly affects the level and precision of the estimated
     emission factor.

     Using the data and statistical analysis conducted in this
     study, coke-pushing emission factors  have been computed which
     characterize the Ford battery  under study.   Although these
     emission factors are expected  to be representative of a fa-
     cility manufacturing blast furnace  coke and experiencing a
     moderate frequency of green pushes, variation in actual emis-
     sions can be expected among coke batteries  due to variation
     in coke oven maintenance,  type of coal  charged, net coking
     time, coking temperatures, meteorological conditions, and
     battery geometries.

     Emission factors are presented in both  metric and English di-
     mensional units.  In other sections of  this report, English
     units are used because, in the U.S.,  common practice in the
     coke production industry incorporates this  system of units.

     Using a scale that associates  a  greenness rating of "1.0"
     with a "clean" push and "3.0"  with  a  "green" push, a series
     of multi-push particulate  tests, whose  average greenness
     ratings range from 1.4 to  2.4  and average 1.8, have been
     used to calculate the following  emission factor estimates.
     Pushing emissions,  expressed as  "kilograms  per metric ton
     of wet coal  charged," and  parenthetically as "pounds per ton
     of wet coal  charged," and  derived using two different methods
     of estimating fugitive emissions, range from 0.60 to 1.9 (1.2
     to 3.8),  and average about 1.2 (2.3)  for filterable particu-
     late; similarly, emission  factors range from about 0.006 to
     0.07 (0.012  to 0.14) and average about  0.033 (0.067) for con-
     densibles.   Of these total pushing  emissions, filterable
     fugitive emissions  are estimated to range from 0.13 to 1.3
     (0.25 to 2.5), and  average 0.60  (1.2);  similarly, the fugi-
     tive emissions of condensibles,  based on the two measurement
     techniques,  are estimated  to range  from 0.001 to 0.05 (0.002

-------
                         - 3 -
the scrubber, where concentrations would be expected to be
the greatest.  Of those materials detected, the following
contaminants have been considered in evaluating the ability
of the control system to both capture and remove them:  sul-
fur dioxide, sulfur trioxide, total sulfur oxides, carbon
monoxide, nitrogen oxides, benzene and homologues, total
light hydrocarbons, methane and homologues-, and ethylene and
homologues.  For many of these materials, total system abate-
ment efficiency varies widely.  This is largely attributable
to emission measurements performed in exhaust airstreams con-
taining very low concentrations of these contaminants.  Gener-
ally, average overall system efficiencies, including estimated
hood capture, range from nil to 60 percent for these materials,
Replicate tests were conducted for each contaminant, some of
which were multi-push samples including a variety of push
types ranging from "clean" to "green."

Maintenance records and on-site system observations indicate
that the emission control system is prone to 20-percent down-
time which results especially during periods when a chronic
failure occurs that is difficult to diagnose.  Based upon sys-
tem observation, the redesign of several system components
would likely reduce downtime significantly.

The costs associated with maintenance, including manpower and
replacement parts, show no obvious trend to increase as the
system ages.  Examination of maintenance and cost accounting
records that document system maintenance shows that average
maintenance costs amount to about $0.15 per ton of coke pro-
duced on the 58-oven battery.

The utility requirements can vary widely, depending upon sys-
tem operation.  By design, the system operates in an idle mode
between pushes, depending upon operator discretion.  Once the
coke guide operator establishes alignment with the next oven
.to be pushed and activates the system, fully operational-mode
status is reached, requiring about 2000 HP of fan power.  De-
pending upon scheduling and the number of pusher machines and
quench cars available to a particular battery in a multi-
battery facility, the utility costs may vary substantially.
Based upon prolonged observation at the Ford Motor coking
facility, electrical energy requirements average about 30
KW-HR per ton of coke produced on the 58-oven battery.

In summary, the system, in its current configuration, abated
pushing emissions during the test program with an efficiency
estimated to range from 41 to 66 percent for filterable par-
ticulate, the prime contaminant.  Nevertheless, the control
system eliminated only about one-half of normal-operation
emissions produced by coke pushing at the time of the study.

Although the system was not operated in accordance with its
original design during the testing and evaluation program, we
view these results as both realistic and representative.

-------
                              - 4 -
     Undoubtedly, the harsh environment and occurrence of green
     pushes,  somewhat characteristic of even a new coke battery,
     are responsible for many of the system modifications that
     have resulted in performance somewhat less than that asserted
     in the Koppers Design Manual.   In our opinion, the mobile
     hood,  as designed, would not efficiently capture pushing
     emissions resultant from a moderately green or green push.
     We believe that capture efficiency can be improved measurably,
     however, if the hood dimensions and exhaust volume rate are
     increased substantially.


1.2  Emission Factors

     All emission factor estimates  contained herein that charac-
     terize total push emissions are necessarily based upon and
     limited  by the technique used  to estimate fugitive emissions.
     The degree to which the fugitive emission estimate is biased
     directly affects the level and precision of the estimated
     emission factor.

     Using  the data and statistical analysis conducted in this
     study, coke-pushing emission factors have been computed which
     characterize the Ford battery  under study.  Although these
     emission factors are expected  to be representative of a fa-
     cility manufacturing blast furnace coke and experiencing a
     moderate frequency of green pushes, variation in actual emis-
     sions  can be expected among coke batteries due to variation
     in coke  oven maintenance,  type of coal charged, net coking
     time,  coking temperatures, meteorological conditions, and
     battery  geometries.

     Emission factors are presented in both metric and English di-
     mensional units.  In other sections of this report, English
     units  are used because, in the U.S., common practice in the
     coke production industry incorporates this system of units.

     Using  a  scale that associates  a greenness rating of "1.0"
     with a "clean" push and "3.0"  with a "green" push, a series
     of multi-push particulate  tests,  whose average greenness
     ratings  range from 1.4 to  2.4  and average 1.8, have been
     used to  calculate the following emission factor estimates.
     Pushing  emissions, expressed as "kilograms per metric ton
     of wet coal charged," and  parenthetically as "pounds per ton
     of wet coal charged," and  derived using two different methods
     of estimating fugitive emissions, range from 0.60 to 1.9 (1.2
     to 3.8), and average about 1,2 (2.3) for filterable particu-
     late;  similarly, emission  factors range from about 0.006 to
     0.07 (0.012 to 0.14) and average  about 0.033 (0.067) for con-
     densibles.   Of these total pushing emissions, filterable
     fugitive emissions are estimated  to range from 0.13 to 1.3
     (0.25  to 2.5),  and average 0.60 (1.2); similarly, the fugi-
     tive emissions of condensibles,  based on the two measurement
     techniques, are estimated  to range from 0.001 to 0.05 (0.002

-------
                         - 5 -
to 0.10), with an average of about 0.018 (0.035).  Of those
pushing emissions captured at the collection hood, the inlet
emissions to the scrubber, expressed as "kilograms per metric
ton of wet coal charged," and parenthetically as "pounds per
ton of wet coal charged," range from 0.48 to 0.65 (0.95 to
1.3) with an average of 0.55 (1.1) for filterable particu-
late; inlet emissions of condensibles measured in the inlet
duct range from 0.0047 to 0.023 (0.0093 to 0.046) with an
average of 0.016 (0.032).  For the control efficiencies
measured, the scrubber outlet emissions of filterable par-
ticulate range from 0.0023 to 0.005 (0.0045 to 0.010) and
average 0.0039 (0.0078).  Condensible emissions from the
scrubber outlet range from 0.0021 to 0.017 (0.0041 to 0.033)
and average 0.008 (0.015).

The aforementioned emission factors are based upon metric
tons of wet coal charged.  Table 1.2 provides the necessary
conversion factors to enable computation of emission factors
in terms of dry coal charged or in terms of coke produced.

As indicated, the particulate emissions, whether fugitive
from the hood collection area or emitted from the scrubber
outlet stack, when compared with the computed pushing emis-
sions, translate to total system abatement efficiencies which
average from about 40 percent to 65 percent.   Conversely,
about 35 to 60 percent of particulate pushing emissions are
uncontrolled, based upon the tests conducted in this study,
including the technique for estimating fugitive emissions.

Emission factors are also developed in this report for certain
gaseous contaminants.  These emission factors are likewise
based upon, and limited by, the methods used to estimate fugi-
tive emissions.  The extent to which estimates of hood capture
efficiency are subject to error affects directly the accuracy
of the emission factors developed for gaseous constituents
(see Section 5.2.3).  In general,  the estimated emission fac-
tors for gaseous components are low when compared to the par-
ticulate emissions.   Particulate apparently constitutes more
than 80 percent of all the pushing emissions.

None of the emission factor estimates presented include door
leaks which, by design, are not controlled by the "Koppers
Smokeless Coke Pushing System."  In that sense, .the results
of this study cannot be compared directly with other coke
oven emission studies which include observations and measure-
ments on total coke-side emissions.

-------
                              - 6 -
                        2.0   INTRODUCTION
The passage and amendment of the Clean Air Act and the continuing
promulgation of emission regulations by federal, state, and local
agencies lead toward specificity in emission restrictions and
control measures for major stationary sources. • In an attempt to
minimize the economic impact of regulatory compliance, the U.S.
Environmental Protection Agency has commissioned a host of.stud-
ies which attempt, as their end purpose, to document the best
practical control technology applicable to specific major emis-
sion sources in various industries.

In the iron and steel industry, metallurgical coke is used as the
primary fuel and reducing agent in the smelting of iron ore in a
blast furnace.  In an integrated steel-making complex, the pro-
duction of coke has been a major source of fugitive emissions
which detract from the quality of the ambient air.

In a step-wise approach to bring steel plants into compliance
with air quality regulations, the U.S. Environmental Protection
Agency has undertaken several studies to define the state-of-the-
art of emissions control from coke production operations, includ-
ing that of "coke pushing."  To abate coke-pushing emissions,
which include visible particulate, hydrocarbons, carbon monoxide,
and a wide variety of organic compounds resulting from incom-
pletely carburized coke, several types of basic emission control
strategies are available.  One such design uses a traveling hood
which captures and conducts coke-pushing emissions into ductwork
leading to a high-energy venturi scrubber.  Long-term performance,
capture of pushing emissions, and scrubbing efficiency of cap-
tured emissions using such a device is the subject of this study.
This report summarizes a comprehensive evaluation of the Koppers
coke pushing emissions abatement system at the Rouge Plant Coke
Battery of Ford Motor Company in Dearborn, Michigan.


2.1  Background and Purpose

     According to a recent study designed to determine the cost
     impact of environmental control regulations on the inte-
     grated iron and steel industry, coke production peaked at
     approximately 61.5 million tons in 1974. (1)  According to
     the EPA compilation of air pollution "emission factors,"
     about one pound of particulate and gaseous pollutants is
     emitted per ton of coke during the pushing operation.(*)
     Therefore, about 60 million pounds of gases and particulate
     matter are discharged to the atmosphere annually from the
     pushing operations at coke batteries.  It is the general
     consensus that the pushing operation represents the second
     largest source of emissions in coke oven operations; uncon-
     trolled charging operations represent the largest single
     source.

-------
     Clearly, the efficient abatement of coke-pushing emissions
     would eliminate some 30,000 tons of atmospheric particulate
     and gaseous contaminants annually if a practical control
     method could be demonstrated and then applied.  The purpose
     of this study is to evaluate the efficiency and viability
     of such an existing control system.  This report, in con-
     junction with similar reports which evaluate other control
     strategies designed to abate coke-pushing emissions) con-
     stitutes a data base upon which regulations designed to
     limit emissions from coke batteries can be developed.

     The Steel Division of Ford Motor Company, Dearborn, Michigan
     is under contract with the U.S. Environmental Protection
     Agency to evaluate the Koppers control system for the abate-
     ment of coke-pushing emissions (EPA Contract No. 68-02-0630).
     As a subcontractor to Ford, Clayton Environmental Consult-
     ants, Inc. furnished services for the emissions testing and
     system evaluation.  In addition to Clayton's work, the sys-
     tem designer and manufacturer, the Koppers Company, wrote a
     "Coke Oven Smokeless Pushing System Design Manual" available
     as a published document from the U.S. Environmental Protec-
     tion Agency.


2.2  Project Objectives

     The overall objective of this project was to evaluate the
     Koppers coke-pushing control system with reference to con-
     trol efficiency, reliability, and cost.   The scope of the
     evaluation included acquisition of historical performance
     and cost data, determination of the nature and extent of
     contaminant species present before and after the abatement
     controls, and compilation of daily records of coke produc-
     tion and system performance, which include documentation of
     causes and costs associated with system downtime.  To pro-
     vide such an evaluation,  the system study was designed to
     achieve the following objectives.

          1.  The primary objective was to evaluate the control
              system's effectiveness for controlling coke-pushing
              emissions by documenting — subjectively and ob-
              jectively—the  control system's efficiency of
              abatement for various contaminant species that are
              characteristic of the pushing operation.  To accom-
              plish the evaluation meaningfully,  it was essential
              to not only quantify scrubbing  efficiency, but also
              measure hood capture efficiency.             ;

          2.  A secondary objective was to maintain records of
              coking time and  production rates as they relate to
              the evaluation of system performance and emission
              factors.

-------
                         -  8 -
     3.  Another objective was to maintain records of the
         system's operability through long-term observation
         with regard to system maintenance and percentage of
         pushes for which the control system functions effi-
         ciently.

     4.  An additional objective was to determine the costs
         associated with operation and maintenance of the
         control system.

     5.  A final objective was to extrapolate the system's
         applicability to new or existing coke oven batteries
         in other locations.

Because the available literature characterizing pushing emis-
sions is inadequate, it was necessary during this project to
determine what contaminant species are present to a measur-
able degree at the inlet to the control system at the Ford
plant.   By documenting the presence of certain contaminant
species, together with characteristic concentrations, a more
credible roster of contaminants emitted from coke pushing is
now available.

In an effort to execute this  project and present the results
in an orderly manner, the project was divided into the fol-
lowing  tasks.
2.2.1  Documentation of Historical Performance

       The objective of this task was to procure the histor-
       ical data which document the operability and mainte-
       nance requirements of the coke-pushing emission con-
       trol system.   Such information is useful in defining
       many of the system malfunctions.   Further,  documenting
       trends in the historical downtime record would reveal
       any tendency  of the system to degrade with  age or use.
2.2.2  Documentation of Historical Economics

       The objective of this task was to procure cost data
       related to the operation and maintenance of the con-
       trol system.   The Ford Motor Company accounts docu-
       ment the funds expended with respect to maintenance
       man-hours and replacement parts.   Analysis of this
       cost data is  useful in determining any tendencies
       toward increased expenditure with time to maintain
       efficient operation of the control system.  Further,
       such data, when normalized to coke production rates,
       are used in defining expected operating costs when
       applying this system to coke batteries in other loca-
       tions.

-------
                         - 9 -
2.2.3  Determination of System Operability

       The objective of this task was to update the recorded
       performance data of the control system by on-site sys-
       tem observation.  The on-site engineer-observer re-
       corded all required process data throughout the course
       of any source test conducted at the inlet and outlet
       of the control system.

       Pertinent process parameters include coal charge
       records, the number and quality of pushes, the coking
       time, the general efficiency of operation of the con-
       trol system as it can be determined by visual obser-
       vations, equipment malfunctions, and necessary cor-
       rective measures.
2.2.4  Measurement of Scrubber Performance

       The objective of this task was to provide a complete
       inventory of the emissions resulting from coke push-
       ing and the scrubbing efficiency for each contaminant.
       Fulfilling the objective requires development and
       testing of acceptable sampling and analytical proce-
       dures for each contaminant.

       Those parameters which are required to correlate
       scrubber efficiency to process conditions include:
       static pressures measured at inlet and outlet of the
       scrubber throat, the scrubber outlet,  the mist elim-
       inator outlet, and the fan inlet to determine the
       contact power transmitted to the gas stream; tempera-
       tures measured in the inlet scrubber water, the inlet
       gas stream, and the outlet gas stream to determine
       vapor-liquid equilibria applicable to  the scrubbing
       of gaseous contaminants; psychrometric calculations
       to determine total water requirements; and measured
       flowrates of the exhaust gas and scrubbing water to
       determine gas-to-liquid ratios for scrubber contact
       power correlations.


2.2.5  Evaluation of the System and Its Applicability

       The objeptive of this task was to assemble an evalua-
       tion which can be used by regulatory officials and
       other coke plant personnel to assess the applicability,
       efficiency, and cost of the mobile hood  coke-pushing
       system.  The degree of emissions abatement from the
       pushing operations and the reliability of the system
       over a prolonged period are criteria used in the
       evaluation,,

-------
                             - 10 -

                                       I
       3.0  COKE PRODUCTION IN THE IRON AND STEEL INDUSTRY
                    AND AT FORD MOTOR COMPANY
Coking is a process by which coal is destructively distilled in
an atmosphere of low oxygen content to'produce volatile gases and
a residue of relatively non-volatile coke.  Of the two processes
available for the manufacture of coke, i the "by-product process"
accounts for more than 90 percent of the coke produced; the bee-
hive process accounts for the remainder.  In the by-product proc-
ess, gases are recovered as they are produced during the coking
cycle.                                 |

A contiguous series of coke ovens, separated by heating flues
placed between ovens, constitutes a "battery" of ovens.  A typical
by-product coke oven is rectangular, 30 to 40 feet long, from six
to 14 feet or more in height, and 11 to 22 inches wide.C3)  in a
modern coke battery, from 16 to 30 tons of coal are charged to
each oven.  Subsequently, the ports are closed and the tempera-
ture in the ovens is maintained at 2000 to 2100°F for 16 to 20
hours.  Coking begins at the oven walls and proceeds toward the
center of the charge.  At the end of the coking cycle, a ram
operating from the "push" side of the oven forces the coke through
the oven and out the "coke" side of the oven where the incandes-
cent coke passes through a temporarily-aligned coke guide and
falls into a quench car.  In the United States, coke is usually
cooled in a quench tower using water sprays.  Water flows over
the incandescent coke for about one minute, after which the coke
is conveyed to a screening location.

If the coking of a given oven charge is complete, most of the
volatile constituents distilled from the coal charge will have
been removed and piped to adjacent chemical plants where usable
products are subsequently extracted.  The combustible portion of
the coke oven gas is usually recycled to the coke battery to sup-
ply requisite heat.

The two major sources of emissions from the coking operation are
the charging and pushing cycles.  With recent innovations, charg-
ing emissions are controllable to a significant degree.  The de-
velopment of practical control technology to abate pushing emis-
sions is the subject of many recent investigations.  The follow-
ing process description and delineation of process emissions pro-
vide an overview of the pushing operation as it relates to the
investigation summarized in this report.


3.1  Description of the Pushing Process

     3.1.1  General Description

            After 16 to 20 hours, the coke must be removed from
            the ovens and quenched.  To remove and collect the
            coke, traveling pusher cars and coke guides serve an

-------
                         - 11 -
       entire battery of ovens.  Several minutes before a
       push is scheduled, door machines on the push and coke
       sides of the oven loosen and remove the doors which
       span the entire height and width of the oven.  After
       the pusher and coke guide have been aligned, coke is
       removed from the oven by a mechanical ram.  The coke
       travels through the coke guide and falls into the
       quench car which is positioned on the coke side of
       the oven being pushed.

       It is necessary for the quench car to move slowly be-
       neath the coke guide to permit even and uniform dis-
       tribution of the coke loaded in the quench car.  When
       the oven is completely empty, the pusher ram withdraws
       and the quench car moves to the quench station, usu-
       ally located at the end of the coke oven battery.

       After bench cleanup, both the push-side and coke-side
       door machines replace the oven doors in preparation
       for coal charging by the top-side larry car.  The
       cycle is then repeated with the oven not being pushed
       again for 16 to 20 hours.
3.1.2  Pushing Process at Ford Motor Company

       Four coke oven batteries produce coke suitable for
       blast furnace use at the Ford Motor Company Steel
       Division, Dearborn, Michigan.  "A" Battery at Ford
       is outfitted with the subject pushing emission con-
       trol device.  The "AX" Battery, an extension of "A"
       Battery, is also serviced by the control system.   The
       10-year-old "A" Battery consists of 45 ovens each 13
       feet high.  Each oven produces, nominally, 12 tons of
       blast furnace coke with gross coking time of 18 hours.
       A "push" is scheduled about every 15 minutes.  The
       "AX" Battery consists of an additional 13 ovens which
       are 13 feet high with each oven capable of producing
       12 tons of blast furnace coke.

       One larry car, one pusher car,  and one coke guide
       serve the 58 ovens in "A" and "AX" Batteries.  The
       quench car serving "A" and "AX" Batteries serves  "B
       Battery as well.
II r> II
       In the typical sequence of operations observed, the
       door machine on the pusher car removes the door of
       the oven to be pushed.   The pusher car.then performs
       the leveling operations for an oven which has just
       been charged by the overhead larry car.   The leveling
       operation takes from two to five minutes to complete.
       The pusher car then returns to the oven  scheduled for
       pushing whose door has  been removed.   During this se-
       quence of operations,  the coke-side door machine

-------
                              - 12 -
            removes the door on the coke side and positions the
            coke guide for the oven scheduled to be pushed.  Both
            machines then await the arrival of the quench car
            which, as previously mentioned, is shared among "A,"
            "AX," and "B" Batteries.  In this sequence of opera-
            tions, doors remain off on both the push and coke
            side for time intervals which range'from two to 15
            minutes, exact times depending on the production
            schedules and the service needs of the two batteries.
            If the operations on either "A" or "B" Battery are
            behind schedule due to any of a variety of reasons,
            from 15 to 45 minutes may elapse before the quench
            car arrives to receive the coke from the oven to be
            pushed.

            In this study, pushing emissions are defined as emis-
            sions leaving the open coke-side door during the 1-
            minute time interval that begins when the pusher ram
            enters the oven.  All coke pushes observed on "A" and
            "AX" Batteries at Ford required from 30 to 45 seconds
            to load the quench car.  Approximately one minute is
            required from the beginning of the coke-push cycle
            for the emission control system to capture the emis-
            sions thus produced.  After the push, the quench car
            travels from the oven just pushed to the quench sta-
            tion located at the south end of "A" Battery.

            Oven No. 1 is nearest the quench station, and Oven
            No. 64 on "AX" Battery is the farthest.  (Oven num-
            bers ending with zero are not included in the oven
            numbering scheme for the 58 ovens.)  Travel distances
            to the quench station for the quench car range from
            about 50 feet to the nearest oven to approximately
            250 feet to the farthest oven.
3.2  Push ing Emi s s ions

     When coke is pushed from a coke oven, relatively fine par-
     ticulate is emitted due to convective currents which entrain
     the loose dust.   Coarser, abraded coke is also produced as
     the hot coke falls into the quench car.  Incompletely coked
     material results in emissions of particulate and volatile
     constituents.  Most of the incompletely coked material usu-
     ally resides near the ends (that is, adjacent to the coke-
     side and push-side doors) where heat losses are the greatest,
     When the oven doors are removed, some of the volatile con-
     stituents of partially-coked material are dispersed to the
     ambient atmosphere.  Depending upon the extent to which the
     end material is  uncoked, emissions can include any and all
     of the by-products produced upon destructive distillation of
     the coal throughout the  entire  coking  cycle.   Reportedly,
     the  emissions  resulting  from uncoked material can include
     ammonia,  coke  oven  gas,  tar,  phenol, light  oil,

-------
                        -  13  -
benzene, toluene, xylene, pyridine as well  as  coal  tar pitch
volatiles and some carcinogenic materials such  as benzo(a)-
pyrene.C^)  More commonly, emissions from partially-coked
materials are comprised predominantly of visible particulate,
light hydrocarbons, carbon monoxide, and some  coal  tar pitch
volatiles.

During pushing, the majority of the emissions which escape
into the ambient air from uncontrolled sources  consist of
particulate matter.(4)  These emissions may originate from
incompletely carburized coal that is adjacent  to the cool
oven doors on the ends of the oven and may, therefore, be
more copious at the beginning and end of a  push.  In those
ovens where insufficient temperatures are maintained during
the coking cycle, the center of the oven charge may also
contain incompletely carburized coal which  contributes to
elevated emissions throughout the push.

As the push is occurring, the quench car necessarily moves
to distribute the coke along its entire length.  During this
time fugitive emissions emanate from the coke over  the length
of the quetich car to a degree which is dependent upon the
degree of carburization.  So-called "green" pushes  emit par-
ticulate matter and volatile constituents throughout the
entire push cycle.  As the quench car then  travels  to the
quench station, these emissions continue to evolve  until the
coke is cooled below incandescence.  During the quench cycle
for such green pushes, the resultant emissions  are  entrained
in the steam produced and pass out the top of the quench
tower.

After the doors are replaced and coal is charged by the over-
head larry car, the newly-introduced coal emits considerable
amounts of volatile constituents as it encounters the heated
oven walls.  Generally, depending upon the degree of door
maintenance, this is the time period in which the largest
amounts of "door leaks" occur as the volume of volatile con-
stituents generated cannot be handled by the by-product col-
lection devices.  These door leaks are at their peak at
charge time and gradually decrease as the coking time in-
creases.  Control of door leaks while the doors are in place
is largely a function of the sdegree of maintenance  practiced.
Additionally, emissions occurring on the push side during
leveling can be responsible for considerable emissions since
they occur at the time when the coal charge is emitting the
largest amount of volatiles.

Typical emissions for coke oven batteries, as they  are docu-
mented in the Compilation of Air Pollution Emission Factors,(
include particulate,  carbon monoxide, hydrocarbons, and am-
monia.  These emissions have been documented to occur during
the coking cycle (as  door leaks) and during the pushing'cycle.
Emission factors for  each of these species are displayed in
Table 3.2-1.  As indicated, depending upon the degree of

-------
                        - 14 -
carburization, other volatile constituents may be emitted.
Table 3.2-2 summarizes those constituents which comprise the
contaminants roster investigated in this study.  Many of
these materials were included as potential process emissions
since they are present in coke oven gas obtained from by-
product recovery.  Experience and the literature indicate
potential emissions, then, of:  nitrogen oxides, sulfur
oxides, hydrogen sulfide, carbon monoxide, methane and its
homologues, ethylene and its homologues, acetylene, total
light hydrocarbons, total reduced sulfur, benzene and its
homologues, pyridine, carbon disulfide, coal tar pitch vola-
tiles, naphthalene, acrolein, phenol, hydrogen cyanide,
chlorine, and ammonia, as well as particulate matter consist-
ing of carbonaceous materials, condensible oils, and various
metallic compounds.

-------
                             - 15 -
              4.0  DESCRIPTION OF THE FORD/KOPPERS
              COKE-PUSHING EMISSION CONTROL SYSTEM
This section describes the design, operating parameters, and
operating characteristics of the emission control system, manu-
factured by Koppers Company, designed to abate coke-pushing emis-
sions.  While the specific parameters which will be listed per-
tain to the Ford Motor Company, Rouge Plant Coke Operation, suf-
ficient data are provided to allow some extrapolation of the de-
sign and operating parameters associated with application of
this system to coke oven batteries in other plants.


4.1  System Design

     In fulfillment of this contract, Koppers Company wrote a de-
     sign manual for its control system as installed at Ford.
     This manual documents the total system as it was designed —
     coke oven pushing emission control capability, operability,
     reliability, and maintainability.   The complete manual is
     available from the U.S. Environmental Protection Agency.
     Salient features are delineated here to provide an overview
     of those portions of the basic control system design that
     were observed to affect materially the operability or main-
     tainability of the system.

     The conventional coke guide and door machine is modified in
     this system to accommodate those appurtenances necessary to
     convey coke emissions into a collection hood.  The hood is
     an integral part of the coke guide.  By providing sufficient
     suction via an induced draft fan upstream and by reducing
     clearances between the hood perimeter and the quench car,
     relatively high indraft design velocities can be maintained.

     Prior to a push, the coke guide is aligned with the oven in
     a conventional manner and the motorized "Rayco" cylinder
     moves an extendible arm located at the top of the hood.  This
     arm opens a set of double doors on an overhead exhaust duct
     which is connected to the induced  draft fan.  As the extend-
     ible arm moves out and contacts two protruding levers, the
     doors open into the duct leading from atop the hood snorkle.
     After reaching its full extension, a limit switch is tripped
     to initiate louver movement upstream of the draft fan, thus
     providing full suction to the exhaust hood.  This circuitry
     is necessary to prevent full suction while the doors are
     being opened.  The extendible arm  would be incapable of open-
     ing the exhaust duct doors against such high negative pres-
     sures.

     The coke, as it is pushed from the oven, falls into the
     quench car whose open top is partially covered by the exhaust
     hood.  As the quench car slowly moves to provide a leveled
     load of hot coke, emissions from the car are drawn into the
     hood and pass through the ductwork leading to the scrubber.

-------
                             - 16 -
     Captured emissions are conveyed through the exhaust ductwork
     and enter the high-energy Kinpactor® scrubber.  The wet
     scrubber is equipped with a recirculated water supply system*
     After being scrubbed, the exhaust gases and entrained mois-
     ture are forced to make a right angle turn.  Larger drops of
     moisture impinge on the bottom of the elbow and provide a
     wetted surface which minimizes abrasive wear.  The gases then
     enter a cyclonic-type mist eliminator.  The water collected
     in this mist eliminator is returned to the water recircula-
     tion system.  The gases pass out of the mist eliminator and
     into an induced draft fan which exhausts to a stack.

     During observation of the control system by the engineeer-
     observer from Clayton Environmental Consultants, Inc., cer-
     tain operational characteristics of the system were noted
     which differ from the system design as it is documented in
     the design manual.  These are described here so that the
     system may be properly characterized when attempting to re-
     late system performance to system design.  Some of these
     changes are further documented in the Koppers design manual
     that delineates "Post Start-up Modifications."

     Due to the modifications, it was not possible to evaluate
     the control system strictly according to original design.
     Nevertheless, because these post-start-up modifications were
     performed out of necessity to provide a functional control
     system within the harsh environment characteristic of coke
     pushing, we believe the evaluation set forth in this study
     represents the Koppers control system as it would exist and
     perform on other coke facilities.


4.2  System Modifications and Observed Operation

     In the original design, the collector hood was fitted with
     movable seal plates designed to move into place and fit over
     the contour of the quench car.  Each seal plate was raised
     and lowered by an electrically actuated cylinder.  In the
     "up" position, normally seven inches of clearance exists
     between the quench car and the locomotive.  In the "down"
     position, approximately two inches of clearance exists.   As
     indicated in the design manual, the operator did not always
     use these seal plates during each push.   After one year, the
     plates were never used.  Therefore, the original plates  were
     removed and fixed plates were installed allowing sufficient
     clearance for the quench car and the locomotive to pass.
     While this afforded some improvement by providing closer
     clearances between the hood and the quench car, air infil-
     tration velocities in the clearances were somewhat less  than
     in the original design.  In addition, it was necessary to
     remove the baffle plates from the quench car and to cut  back
     the end plates of the quench car which further reduced in-
     filtration velocity,  a crucial parameter that governs the
     efficiency of emission capture around the hood perimeter.

-------
                        - 17 -
As noted in the design manual, the synchronization unit al-
lows the operator to provide a level load of coke in the
quench car without seeing the coke in the quench car (due to
the built-up ends designed to maximize indraft velocity).
However, critical alignment problems coupled with airborne
dust resulted in the rapid deterioration of the synchroniza-
tion unit.  After several attempts to repair the unit, the
synchronization unit failed to perform adequately.  As a re-
sult, it was necessary to remove enough of the end plates on
the operator's side of the quench car to permit visual level-
ing of the coke.

The design manual describes a motorized cylinder that extends
the telescoping section of the hood snorkel to connect to the
exhaust duct.  The control system was provided with a butter-
fly damper designed to open and relieve suction in the ex-
haust duct so that the doors (which are accessed by the hood
snorkel) would be easier to open.  Koppers Company conducted
a test in 1973 to determine the extent of air iiileakage, and
therefore of negative pressure in the exhaust duct, when the
butterfly damper was not used.  Apparently, sufficient leak-
age existed around the multitude of doors in the duct to re-
lieve the suction and assure that the electric cylinders
could open the doors even with the butterfly damper closed.
Since these tests indicated that the butterfly damper was
not required, it was welded shut.  In other sections of this
report (those specifically related to 'maintenance and repair
requirements) it can be noted that the motorized cylinder
(called the Rayco cylinder) is at times incapable of opening
the doors without drawing excessive amperage thus tripping a
circuit breaker.  Such a malfunction disables the control
system, rendering it ineffective.  While several factors af-
fect the ability of the Rayco cylinder to open the exhaust
duct doors, certainly the added resistance caused by the
negative pressure in the duct with the butterfly damper
welded shut contributes to the electrical overload.

At the Ford coking operations, several quench cars are used
to service both "A" and "B" Batteries.  Maintenance require-
ments necessitate that some quench cars without built-up
ends and sides be used on the "A" and "AX" Batteries in con-
junction with the control system.  In the future, as other
coke oven batteries at Ford are equipped with similar control
systems,  other quench cars will likely be modified.  However,
it should be noted that during some of the sampling conducted
and reported the quench cars without built-up sides and end
plates were in use.  Records are not complete enough to indi-
cate the sampling days when the unmodified quench cars were
in use.  However, indications are that the modified quench
car was in use during more than 90 percent of the testing
and evaluation program.

-------
                        - 18 -
The emission control system, as indicated in the Koppers
manual, was designed to service coke batteries which produce
a minimum number of green coke pushes.  When a large volume
of combustible gases and products of combustion evolve during
pushing, indraft velocities are insufficient to contain the
fugitive emissions.  This problem is compounded by the fact
that the hood extends over only 30 feet of the 40-foot-long
quench car.  As the quench car necessarily moves under the
hood to provide a level load of hot coke, a green push will
continue to generate fugitive emissions in those sections of
the car which are beyond the edge of the hood.  In addition,
although the hood is 30 feet long, the rectangular opening
at the base of the hood is not 30 feet long.  At the end of
the actual hood opening, flat, horizontal plates provide a
path for transporting emissions from the solid-plate portion
of the hood into the suction area of the hood.  When the baf-
fle plates were removed, however, this solid plate area be-
came  largely ineffective in capturing emissions generated
under it.  Consequently, only about 50 percent of the quench
car is under the effective hood area at any one time.  Dur-
ing clean pushes, this is sufficient to capture much of the
fugitive emission.  In green pushes, fugitive emissions in-
crease markedly.  The estimated extent of this fugitive
emission increase is presented in other sections of this
report (see Section 6.4).

As reported in the design manual, exhaust gases discharging
from the fan outlet pass directly into the base of the dis-
charge stack.  This configuration caused a noise problem at
the outlet stack.  To attenuate the noise, the airflow against
the outer wall at the base of the stack was minimized by
placing brick in the stack base to form a radius to turn the
airflow.   This decreased the noise and vibration levels
measurably.  To initially reduce the stack noise level, an
acoustical silencer was installed in the outlet stack.  The
stack silencer consists of plates extending along the center-
line of the stack in such a way as to divide the stack into
two semicircles.  The stack silencer extends from several
feet above the top of the bricks at the base of the stack to
the top of the stack.  Before testing could be conducted for
this project, it was necessary to install a stack extension
that provided approximately 30 feet between the top of the
stack and the top of the stack silencer.   This modification
was for test purposes only.

Again,  these sys.tem modifications are evidently a necessary
compromise to the original design configurations so that a
reliably  functional system could be ensured.   As such, al-
though tested in a modified  state,  this program provides an
evaluation of a control system as it would likely exist on
another battery of coke ovens and the harsh environment
present there.

-------
                             - 19 -
4.3  Operating Parameters

     To provide the requisite air movement and capture efficiency
     at the hood, the induced draft fan was designed to provide a
     gas flowrate through the system of 112,000 ACFM at a nominal
     gas temperature of 132°F and a system static pressure less
     than -60 inches of water.   The fan is driven by a 2,000-HP,
     3-phase motor drawing approximately 240 amperes at 4160
     volts when the fan inlet louvers are in the full open posi-
     tion.  Between pushes,  when the louvers are closed, the
     emission control system was designed to require approximately
     1000 HP.  Presented in  other sections of this report are the
     actual exhaust volumes, temperatures, pressures, and utility
     and power requirements  for actual operating conditions.  Re-
     petitive flow measurements indicate that the actual exhaust
     volume during a coke push  is somewhat lower than the design
     flowrate.  Further, inlet  gas temperatures vary widely de-
     pending upon the degree of greenness of the coke push — rang-
     ing typically from 100°F to over 700°F.

     Calculations indicate that when no coke was being pushed into
     the quench car, the design flowrate would provide an indraft
     velocity of approximately  30 feet per second at the hood
     perimeter if the originally designed seal plates were in
     place.  Settling and movement of the quench tracks and re-
     moval of the side seal  plates have reduced indraft veloc-
     ities to approximately  10  feet per second.  Depending upon
     ambient weather conditions and the greenness of coke pushed,
     these indraft velocities were not always sufficient to cap-
     ture the fugitive emissions efficiently.  Capture efficien-
     cies are documented in  other sections of this report.

     The venturi scrubber was designed to operate at a pressure
     differential of 60 inches  of water.  In practice, pressure
     differentials measured  across the scrubber throat ranged
     from about 30 to 45 inches of water.  This differential
     pressure drop existed both prior to and after replacement of
     all scrubber throat nozzles as documented by Ford maintenance
     personnel.  This suggests  that water spray coverage of the
     throat cross section is likely not responsible for its dif-
     ferential pressure being less than the design pressure drop.

     As indicated, sound attenuation devices and flow alignment
     bricks in the outlet stack were installed.  Undoubtedly,
     these devices provide a restriction to gas flow not antici-
     pated in the original design.  This is evident since the
     system was designed to  operate at a pressure of -65 inches
     of water at the mist eliminator outlet (just upstream of the
     fan inlet), while, typically, about -48 to -55 inches of
     water pressure was measured at the mist eliminator outlet.

     The scrubber requires approximately 890 gpm of recirculated
     water according to design  criteria.  Measured water flow-
     rates were approximately at this value during the testing

-------
                        - 20 -
documented in this report.  However, historical operating
data measured by the flow recording devices in the control
room indicate that this flowrate varies considerably.  These
data indicate that water flow, determined as monthly averages,
varies from about 600 to 1000 gpm.

The design manual presents calculations of operating costs
and utility requirements based upon design data and a pre-
scribed schedule which the system operator should follow.
However, our observations show that the operator does not
always operate the system in accordance with design cri-
teria.  During a typical 1-hour period of operation at the
Ford Motor Company "A" and "AX" coke batteries, there are
approximately four coke pushes.  Design specifications men-
tion that the collection hoods should be put in place approx-
imately  two  minutes before the coke push.  Since the coke
push lasts approximately 40 seconds, utility requirements
set forth in the design manual were based on two and one-
half minutes of system operating per push or 10 minutes of
full system operation during each push-hour.  In actual
practice, the operator is ready for the next push, including
full activation of the emission control system, from two to
15 minutes before the push occurs.  As previously indicated,
because the quench car is shared among "A," "AX," and "B"
Batteries, the coke guide operator on "A" and "AX" Batteries
sometimes has a substantial interlude between coke pushes as
determined by "A" and "B" coke battery scheduling.  The ef-
fect of actual plant operation on utility costs is described
in other sections of this report.  However, it is noted that
the total operating cost of this emission control system is
affected materially by the production schedules and mainte-
nance requirements at a particular multi-battery coke-making
facility.  Operator discretion is very important in deter-
mining actual operating costs.

The values of the operating parameters listed above are, in
many cases, different from design values.  Section 6.1.1
addresses the range of operating parameters encountered dur-
ing the testing and evaluation program as they compare to
average values computed from historical data recorded
throughout the control system's life.  The data indicate
that the values of operating parameters listed above are
characteristic of the control system's typical operation.

-------
                             - 21 -


                    5.0  FIELD INVESTIGATION
This section describes the methodology of acquiring emission
measurement data and process parameters.  Prior to beginning this
study, it was necessary to define pertinent operational param-
eters which must be monitored or measured to characterize both
process and control system conditions.  After the definition of
these parameters, methods of measurement and data acquisition
were specified to ensure that the necessary information could be
obtained in the most accurate manner possible.

To determine what samples had to be acquired to characterize
emission rates and collection efficiencies, the nature of the
emissions was defined by literature investigation and range-
finding tests at the scrubber inlet where concentrations are at
a maximum.  For each emission species detected, appropriate sam-
pling and analytical methods at the scrubber inlet and outlet
were chosen to assure accurate assessment of scrubber efficien-
cies for the control system.

From the earliest phases of this investigation, it was obvious
that any complete and meaningful evaluation of the Ford/Koppers
coke-pushing control system must include an analysis of the un-
captured coke-pushing emissions which appeared to be a signifi-
cant portion of the total push particulate emissions.  However,
to characterize the extent of fugitive emissions into the am-
bient air at the hood inlet, it was necessary to utilize a sam-
pling technique which provides a good estimate of hood capture
efficiency.

Proven and accepted methods for quantifying fugitive emissions
do not exist, although the state-of-the-art continues to develop
steadily.  Because of the visual appearance of uncaptured push
emissions and the availability of large numbers of pushes from
which to judge the degree of capture of the control system, it
was decided that a subjective estimate of effectiveness of hood
capture would be an important component in the analysis of over-
all system performance.

The estimated hood capture efficiency, applied to the appropriate
techniques to assess accurately the measured inlet emissions, pro-
vides an estimate of the mass rate of fugitive emissions.  From
this estimate and the measured scrubber outlet emissions, a range
of total, overall efficiencies of the control system could be
calculated to determine an estimate of total abatement efficiency
of the coke-pushing emissions.  The test results were then nor-
malized with respect to process performance characteristics so
the information could be extrapolated to similar coke oven bat-
teries in other plants.  Consequently, it was necessary to define
process conditions and develop efficient methods to acquire all
necessary process data for each push measured during the study.

-------
                             - 22 -
5.1  Methodology of Data Acquisition

     In fulfilling the project objectives it was necessary to:
     1) document historical performance, 2) document historical
     economics, 3) determine system operability, 4) measure
     scrubber performance, and 5) evaluate the emission control
     system and its applicability to other coke oven batteries.


     5.1.1  Documentation of Historical Performance

            The emission control system was installed in June,
            1972.  Therefore, it was necessary to evaluate written
            records which characterize the operating and mainte-
            nance requirements for the emission control system.
            These data were then updated by an engineer-observer
            from Clayton Environmental Consultants, Inc., who
            acquired field data intermittently between August,
            1974 and August, 1975.

            Correlations (developed in latter sections of this
            report) relate overall scrubber efficiency and hood
            capture efficiencies to certain design parameters
            which are recorded routinely by coke oven personnel.
            Written records included:   1) "Operating Checks on
            Smokeless Coke Pushing Emission Control System Once
            Each 8-hour Shift by the Coke Battery Foreman,"
            2) " 'A1 Coke Battery Door  Machine Downtime," 3) "Opera-
            ting Checks on the Smokeless Coke Pushing Emission
            Control System Once a Week on Saturday by the Second
            Shift General Foreman," 4) "Maintenance Department
            Logbook," and 5) "Operations Department Logbook."

            The written records of the daily operating checks con-
            ducted by the coke battery foreman contain:  fan bear-
            ing temperatures; motor bearing temperatures; fan am-
            perage requirements before and during a coke push; gas
            temperatures at the exhaust duct and Kinpactor inlet,
            fan inlet,  and the fan discharge stack; Kinpactor dif-
            ferential pressure drop both between and during coke
            pushes; water flowrate to  the Kinpactor; pressure at
            the water pump and which of the two pumps is operating;
            an observation before noon of one coke push to deter-
            mine the stack opacity.   In addition, ancillary notes
            are written at the bottom  of these sheets as deemed
            appropriate by the coke battery foreman to character-
            ize the emission control system performance.  A speci-
            men of this document is provided in Figure 5.1.1-1.
            These data  are used to calculate utility requirements
            as they actually occurred  throughout the lifetime of
            the emission control system.

            The "A" coke battery door  machine downtime is docu-
            mented in a notebook kept  by the coke oven personnel.
            This record was kept routinely even before installation

-------
                 - 23 -
of the "Smokeless Coke Pushing System."  Subsequent
to modification of the "A" coke battery door machine
necessitated by installation of the emission control
system for "A" and "AX" battery, records were kept
of downtime attributable to the door machine itself
(that is, actual door machine failures independent
of the control system).  In determining actual emis-
sion control system maintenance, these records were
researched for entries pertaining to downtime asso-
ciated only with the "Smokeless Coke Pushing System."
A specimen of this document is included in Figure
5.1.1-2.

The weekly operating checks by the general foreman
document:  oil levels in the fan bearings and pumps,
pump performance, Kinpactor spray header pressures,
and records of when Kinpactor sprays were reamed.
Additionally, ancillary notes are recorded at the
bottom of these sheets which document the emission
control system performance deemed important by the
general foreman.  A specimen of this record is pro-
vided in Figure 5.1.1-3.  These records were researched
for pertinent entries which document unusual perform-
ance of the emission control system and to assure that
proper maintenance was performed routinely as required
by the preventive maintenance schedule.

The maintenance department logbook contains entries
pertaining to all coke batteries serviced by the main-
tenance department.  Even for "A" coke battery, main-
tenance work performed upon the "Smokeless Coke Push-
ing System" is interspersed with regular maintenance
performed on "A" coke battery.  Approximately 16
volumes of the maintenance department logbooks were
researched to extract entries which pertained only to
the "Smokeless Coke Pushing System."  Figure 5.1.1-4
includes a specimen of the typical notes contained in
maintenance logbooks.  Entries pertaining to the coke
pushing emission control system maintenance are pre-
sented completely in Appendix A.

The operations department logbook summarized opera-
tional problems or difficulties associated with "A"
and "B" coke batteries.  The only logbooks available
for perusal which pertained to the emission control
system were entries beginning in March, 1975 and ex-
tending to August, 1975.  A typical specimen of these
entries is shown in Figure 5.1.1-5.  A complete list-
ing of logbook entries pertaining to the emission
control system is contained in Appendix A.

These sources of information constitute all available
documents which can be used to characterize the his-
torical performance during the lifetime of the coke
pushing emission control system.

-------
                        - 24 -


5.1.2  Documentation of Historical Economics

       Any attempt to characterize dollar value expenditures
       throughout the history of the emission control system
       should inherently include costs associated with opera-
       tion of the system;  these are composed of utility
       costs and maintenance costs.   While it is straight-
       forward to obtain utility costs by extrapolation of
       fan and pump horsepower ratings, more accurate util-
       ity requirements and costs can be determined by re-
       searching available  records which document actual
       utility usage.  Similarly, maintenance costs can be
       approximated by attempting to estimate man-hours ex-
       pended and multiplying by the cost per man-hour.
       However,  such an approach leads to erroneous mainte-
       nance costs because  maintenance records are inherently
       designed  to document the causes of downtime and, as
       can be seen by inspection of  maintenance records, do
       not include actual man-hour requirements to effect
       the necessary repairs.

       Consequently, it was necessary to investigate cost
       accounting files and control  codes to determine actual
       maintenance costs (labor and  material) associated with
       the emission control system.   Two methods of cost ac-
       counting  procedures  were investigated to document his-
       torical economics.   The first method  included review
       of  overall cost control codes associated with:   1) coke
       battery pollution control including labor and parts
       for the "A" and "AX" emission control system, and
       2)  pollution control by-products cost of disposal.
       Two 4-digit cost codes  were obtained  from cost  ac-
       counting  personnel and  all computer records of  year-
       to-date totals associated with these  two cost codes
       were researched.

       Upon the  advice of cost accounting personnel that
       these records were conglomerates and  did not neces-
       sarily reflect all costs associated with the system,
       a second  method of cost accounting was researched.
       In  this method, a special detailed 4-digit cost code
       (which is a subcontrol  code of the major cost codes)
       was used  to obtain monthly totals of  "maintenance
       labor requirements"  and "stores payable cost require-
       ments."  These control  codes  were established specif-
       ically by the cost accounting department to provide
       the most  accurate possible measure of costs associated
       with the  emission control system.   The results  of
       these documentary investigations are  displayed  in
       other sections of this  report.

-------
                        - 25 -
5.1.3  Determination of System Operability

       Simultaneous with all testing,  an on-site engineer-
       observer recorded certain key operational variables
       necessary to characterize hood  capture efficiency.
       Figure 5.1.3-1 displays a specimen of the daily log
       of hood performance used to record the engineer-
       observer's data regarding hood  capture.   These data
       include:  the oven number pushed, the scheduled push
       time,  the actual push time, the measured cross-draft
       wind velocity measured in the hood proximity,  the
       estimated hood capture efficiency, the stack opacity,
       the condition of the push as determined  on the coke
       side,  and the net amount of coal charged to the oven
       just pushed.

       Cross-draft wind velocity measurements were made with
       a hand-held vane-axial anemometer facing into  the wind
       and held by the engineer-observer located about 200
       feet west of "A" Battery.

       The hood capture efficiency recorded on  the data sheet
       shown  in Figure 5.1.3-1 was obtained by  visual estima-
       tion of the percentage of emissions emanating  from the
       quench car which were collected by the hood.  Although
       this is a semi-quantitative determination, the degree
       to which this measurement corresponds to another esti-
       mate of fugitive emissions (obtained by  high-volume
       air sampling) is discussed in other sections of this
       report.  In general, any engineer-observer visually
       estimating hood capture efficiency relied upon a
       "conceptual record" of uncontrolled push emissions
       for a  given greenness to judge  and compare the visible
       fugitive emissions to the uncontrolled emissions char-
       acteristic of that type of push.  The opacity  of the
       fugitive emissions is a useful  aid in categorizing  the
       type of push observed and judging the magnitude of  an
       uncontrolled push.

       Figure 5.1.3-2 is a schematic diagram showing  the ap-
       proximate locations of the engineer-observer during
       the push.  The location varied, due to line-of-sight
       restrictions, depending upon the oven pushed.

       Stack  opacity was determined by certified opacity
       readers, in most cases.  While  the majority of stack
       opacity measurements were made  by individuals  who had
       successfully completed an EPA-approved course  in
       opacity reading, all opacity readings were taken by
       individuals who had extensive experience in evaluating
       visible emissions and who were  oriented  and trained
       at the field site by certified  opacity readers.

-------
                        - 26 -
       "Coke-side condition" was a subjective judgment re-
       corded by the engineer-observer to document whether
       the push was "green," "medium," or "clean."  (Clayton
       used these three categories;  Ford personnel used only
       two classifications.)  Net coal charged was obtained
       from process data records.

       The daily log of hood performance was recorded  for
       nearly all times during which actual testing occurred.
       At other times,  the engineer-observer was stationed
       at the fan where he recorded  data pertinent to  the
       scrubber performance evaluation.
5.1.4  Measurement of Scrubber Performance

       To characterize the scrubber performance,  it was neces-
       sary to determine both the scrubber operating param-
       eters and the actual inlet and outlet scrubber emis-
       sion rates.  Emission rate determinations  are discussed
       in Section 5.2 of this report.  To characterize the
       scrubber performance parameters,  the engineer-observer
       recorded data on data sheets of the type displayed in
       Figure 5.1.4.  When tests were not being conducted,
       the engineer-observer was present at the fan to record
       data on the "Daily Log of Scrubber System  Performance."
       In addition,  these scrubber operating conditions were
       also recorded by the sampling crew during  the course
       of performing inlet and outlet source testing.

       As indicated  in Figure 5.1.4, this data  sheet contains
       information which includes:  the  number  of the coke
       oven just pushed; push time; conditions  existing at
       the scrubber  inlet, including dry-bulb temperature,
       wet-bulb temperature, static pressure, and velocity
       pressure; scrubber throat conditions, including pres-
       sure drop, water pressure, and water temperature;
       scrubber outlet conditions, including dry-bulb tempera-
       ture, wet-bulb temperature, and static pressure; recir-
       culated water flowrate to the scrubber;  and the fan
       motor current.  Several weeks of  pretest observations
       established that the scrubber throat conditions, scrub-
       ber outlet conditions, recirculated water  flowrate,
       and fan motor current were quite  constant  during a
       day's testing and even during several days of testing.
       Inlet scrubber conditions varied  somewhat  depending
       upon the "greenness" of the coke  push.  These system
       measurements, as just noted, remained quite constant.
       Therefore, the engineer-observer  recorded  these meas-
       urements only a few times during  the day to adequately
       characterize  the performance of the emission control
       system.  At other times,  the engineer-observer accumu-
       lated hood capture data.

-------
                        - 27 -
       Because the scrubber inlet and outlet are at different
       elevations approximately 35 feet apart, it was not
       possible for the engineer-observer to physically trav-
       erse the distance between the inlet and outlet during
       the 40-second duration of a push.  Consequently, the
       inlet dry-bulb temperature measurement, and the inlet
       static and velocity pressures were measured remotely
       at the scrubber outlet by means of thermocouples con-
       nected to a potentiometer and tubing connected to an
       inclined draft gauge, respectively.  Scrubber outlet
       dry- and wet-bulb temperatures were measured directly
       at the elbow leading from the mist eliminator outlet
       to the fan inlet.  Scrubber outlet static pressure
       was measured at that location using a mercury manom-
       eter.  A differential pressure gauge installed near
       the fan permitted the engineer-observer to record
       scrubber throat pressure drop during a coke push.
       Water pressure was recorded by observing gauges on
       the scrubber pumps located in the pump house near the
       fan.  Water temperature was obtained by opening a
       valve on the inlet water line and allowing the water
       to run for several minutes before taking the water
       temperature using a mercury thermometer.  At the end
       of each day's testing the engineer-observer would
       record the recirculated water flowrate and fan motor
       current from the strip chart recorders located in the
       "A" and "B" coke battery control room.  These record-
       ers are the same recorders used by the Ford personnel
       in filling out their records of system operation dis-
       cussed in Section 5.1.1.
5.1.5  System Evaluation and Applicability

       To normalize all maintenance requirements, maintenance
       and operation costs,  hood collection efficiency, and
       scrubbing efficiency, certain process parameters were
       obtained to characterize the push sampled.  Figures
       5.1.5-1, -2, and -3,  respectively, display specimens
       of the coke oven heater inspection report, coal charge
       log, and oven temperatures,  as they are recorded by
       Ford Motor Company personnel on a routine basis.  These
       documents provide all available data to characterize
       those process parameters which determine, to the extent
       possible, emissions generated during a coke push which
       must be abated by the control system.

       The coke oven heater  inspection report is used to
       determine the time delay from the scheduled to the
       actual coke push.  This report also documents the
       Ford Motor Company quench car operators'  appraisal
       of the coke's condition on the coke side  of the oven
       during the push.  The two designations used by Ford
       personnel include "fire and  smoke" (not analogous to

-------
                             - 28 -


            the fire and smoke designation used by the Clayton
            engineer-observer) and "good."  The latter category
            designates a relatively clean coke push.  These data
            were used to determine the correlation between the
            engineer-observer's and the Ford Motor Company person-
            nel's observations of coke push condition.

            The coal charge log documents the time when a particu-
            lar oven was charged with coal.  The net weight of
            coal charged is obtained from a printing recorder lo-
            cated in the "A" and "B" coke battery control room.
            Each larry car is weighed upon receiving a charge of
            coal which is completely emptied into a coke oven.

            The coke oven temperature data sheet documents tem-
            peratures measured once each shift by Ford Motor Com-
            pany personnel using an optical pyrometer.  Tempera-
            ture measurements are taken at portholes located on
            the roof of each oven on the push and coke sides.  The
            data are needed in attempts to predict the "degree-of-
            greenness" of a push as it relates to coking tempera-
            ture.   Results of the attempted correlation are dis-
            cussed in other sections of this report.


5.2  Emissions Measurement

     At the onset of this project in 1974, a literature investi-
     gation was performed to provide a roster of emissions pro-
     duced by the coking process.  Based upon extensive experience
     with sampling of other coke pushing emission effluents, the
     list of suspect contaminants was later refined and expanded
     to include the total contaminant roster displayed in Table
     3.2,2.  Range-finding tests were then conducted intermittent-
     ly during the first four months of 1975.  From these data,
     average coke pushing emission rates were documented for many
     of the compounds in Table 3.2-2 that were expected to exist
     in low or immeasurable concentrations.  Based upon these
     data, certain contaminants were excluded from the roster for
     subsequent source testing conducted during June and July of
     1975.

     The comprehensive source sampling project thus included both
     the range-finding tests conducted during the first four months
     of 1975 and the further, subsequent testing conducted during
     June and July.  Those contaminants included in this comprehen-
     sive source sampling project were the following:   ammonia,
     hydrogen sulfide, phenolics, sulfur dioxide, sulfur trioxide,
     total sulfur oxides, nitrogen oxides, benzene and homologues,
     pyridine, carbon disulfide,  carbon monoxide, ethylene and
     homologues, total light hydrocarbons, methane and homologues,
     acetylene, naphthalene, fluorene, 9-fluorenone, phenanthrene,
     anthracene, acridine, carbazole, fluoranthene, pyrene, chry-
     sene, benzo(a+e)pyrene, acrolein, chlorine, cyanide, and

-------
                        - 29 -
total reduced sulfur.  In addition, particle sizing tests
and particulate loading tests were conducted.

The following subsections present a description of the sam-
pling and analytical methods used to measure contaminants at
the wet scrubber inlet and outlet sampling locations.  During
range-finding tests, samples were collected only at the in-
let location, and following range-finding tests, sampling for
some contaminants was performed only at the scrubber outlet.
This sampling at the outlet only was carried out to further
document the low or immeasurable concentrations found at the
inlet.  In all cases of scrubber efficiency measurements,
however, sample collection at the outlet was simultaneous
with sample collection at the inlet.

Figure 5.2 indicates that sampling at the inlet location was
conducted through each of two ports in the 66-inch diameter
horizontal duct immediately upstream of the scrubber.  Ap-
proximately 30 feet of straight duct upstream and four feet
downstream of the sampling ports yielded a uniform velocity
profile at this sampling location.

Four ports located 90 degrees apart were positioned at the
90-inch diameter outlet location, as shown in Figure 5.2.
At this location, approximately 23 feet of straight duct
below the sampling ports and seven feet of straight duct
above the sampling ports provided a fairly uniform velocity
profile.

Sampling methods used at both locations followed the prin-
ciples outlined in "Standards of Performance for New Station-
ary Sources,"(5) presented in detail in Appendix B.  The sam-
pling location and marking of sampling points followed a
modified version of EPA Method 1.  This modification involved
using fewer sampling points than the method requires for the
location of the sampling point upstream and downstream of po-
tential disturbances.  Due to the pushing schedule, it was
impracticable to sample at the number of points required.
Even during the course of the actual sampling it was neces-
sary to reduce the number of sampling points from 24 to 16
in order to obtain a particulate sample in a reasonable
length of time.

Velocity traverses were conducted according to EPA Method 2,
and exhaust gas composition and moisture content were deter-
mined according to EPA Methods 3 and 4, respectively.

Because of the intermittent nature of this emission source,
sampling was conducted only during the brief periods of coke
oven pushing.  Typically, the push on "A" Battery at the Ford
Rouge Plant requires about 40 seconds to be completed while
about 60 seconds are required to completely evacuate the con-
trol system and ductwork of push emissions.  All non-grab
samples spanned this one-minute period.  Sampling commenced

-------
                        - 30 -
as coke first fell into the quench car and continued until
either the designated period of sampling had elapsed or un-
til the snorkle was removed.

The field data sheets for all source sampling are presented
in Appendix C.  Calibration data for all Fitot tubes, dry-
gas meters, and critical orifices used in the study are pre-
sented in Appendix D.  Appendices E and F present sample
calculations and project participants, respectively.

Thirty-two emission species were sampled and analyzed during
the source sampling phase of the project; however, somewhat
less than 32 different sampling methods were employed because
aliquots of a collected sample could often be analyzed for
several contaminants.  A description of each of the sampling
methods follows with subsections under each method describing
the analytical methods used for each of the contaminants col-
lected in that sample.
5.2.1  EPA Method 5 Sampling

       The particulate sampling method very closely followed
       that outlined by EPA Methods 1 through 5 (Appendix B).
       Particulate was withdrawn from the stack isokinetically
       through a stainless steel probe to a heated glass-fiber
       filter.  Non-filterable particulate was captured sub-
       sequently in three impingers containing 200 ml of dis-
       tilled water.  The volume of sampled gas was measured
       with a calibrated dry-gas test meter.  Sampling rates
       were established using an orifice meter and inclined
       manometer.  Skilled sampling team members,  having ex-
       tensive experience with the temperature and velocity
       pressure fluctuations characteristic of the push emis-
       sions captured in the inlet ductwork, were  able to at-
       tain excellent isokineticity.

       The percent isokineticity ranged from 100.0 to 108.6
       and averaged 105.4 during the  test series at the inlet.
       At the outlet, percent isokineticity ranged from 102.2
       to 113.2 and averaged 107.3 during the nine particulate
       test series.  Calculation of the percent of isokine-
       ticity on a push-by-push basis yielded similar results.
       Within each one-minute sampling period, conditions of
       velocity pressure and stack temperature varied consi-
       derably. .  Adjustments in sampling rate were made within
       each sampling period as well as between sampling points.
       Rapid adjustment was facilitated by the fact that the
       "fine" setting on the meter console could be adjusted
       during sampling at one point.   Closing the  "coarse"
       valve without adjusting the "fine" valve resulted in
       a similar sampling rate when the coarse valve was
       opened again.  Fortunately, the required flowrate from
       point to point did not vary greatly.   Calculation of

-------
                        - 31 -
       required flowrate was done with a portable electronic
       calculator and was facilitated by charts of the re-
       quired rate at various velocity pressures and stack
       temperatures.
       5.2.1.1  Particulate

                Filterable particulate includes those frac-
                tions captured in the acetone rinse of the
                sampling probe and on the glass-fiber filter.
                Condensibles are determined from the residues
                after evaporation of the impinger solution.

                In the laboratory, the acetone probe wash was
                evaporated to a constant weight at room tem-
                perature and subsequently weighed on an ana-
                lytical balance capable of sensing 0.1 milli-
                gram.  The glass-fiber filter was brought to
                a constant weight in a desiccator and was
                similarly weighed.  Aqueous impinger solutions
                were extracted repeatedly with a mixture of
                chloroform-ether that separated the organic
                materials from the inorganic materials.  Both
                the aqueous and chloroform-ether solutions
                were subsequently dried to a constant weight
                and weighed on an analytical balance.  The
                analysis of the filters and solutions paral-
                leled that of the particulate samples, and
                particulate weights were appropriately ad-
                justed for blank values.
5.2.2  EPA Method 8 Sampling

       The sampling train consisted of a sampling nozzle, a
       glass-lined probe, a heated 110-ml glass-fiber filter,
       a Greenburg-Smith impinger containing 100 ml of 80-
       percent isopropyl alcohol, and two impingers each con-
       taining 100 ml of 3-percent hydrogen peroxide.  The
       impingers and a condensate trap containing silica gel
       were immersed in an ice bath.

       The sampled gas was drawn at approximately one CFM by
       means of a vacuum pump through a calibrated dry-gas
       meter where the temperature and the pressure of the
       gas were measured, then through a calibrated orifice
       meter.  After each test the contents of the first im-
       pinger and that of the second  and third impingers were
       transferred to two separate sample bottles for trans-
       port to the laboratory.  The volume of the metered gas
       was determined from the initial and final readings of
       the gas meter.  The average meter pressure and tem-
       perature were used to calculate the sampled gas volume
       at standard conditions.

-------
                        - 32 -
       5.2.2.1  Sulfur Dioxide
                An aliquot portion of the hydrogen peroxide
                sample solution was transferred to an Erlen-
                meyer flask.  Isopropyl alcohol was added to
                make an 80-percent isopropyl alcohol solution
                and two to four drops of thorin indicator
                were added.  The solution was then titrated
                with standardized barium perchlorate solution
                to a pink endpoint.  The results were re-
                ported as sulfur dioxide.
       5.2.2.2  Sulfur Trioxide
                An aliquot portion of the isopropyl alcohol
                sample solution from the first impinger was
                transferred to an Erlenmeyer flask.  Two to
                four drops of thorin indicator were added.
                The solution was then titrated with stand-
                ardized barium perchlorate solution to a
                pink endpoint.  The results were reported as
                sulfur trioxide.
5.2.3  Sodium Hydroxide Sampling

       The sampling train consisted of a sampling nozzle, a
       glass-lined probe, a heated glass-fiber filter, and
       two Greenburg-Smith impingers,  each containing 150 ml
       of 0.1-N sodium hydroxide.   The impingers and a con-
       densate trap containing silica  gel were immersed in
       an ice bath.

       The sampled gas was drawn at approximately one CFM by
       means of a vacuum pump through  a calibrated dry-gas
       meter where the temperature and the pressure of the
       gas were measured, then through a calibrated orifice
       me t e r.

       After each test, the impinger contents and the rinsings
       of the impingers were transferred to sample bottles for
       transport to the laboratory.  Similarly,  the probe was
       rinsed with the sodium hydroxide solution, and the filter
       was added to the rinsings.   "Filterable"  material was
       captured in this sample fraction.

       The volume of metered gas was determined  for the ini-
       tial and final readings of  the  gas meter.  The average
       meter pressure and temperature  were used  to calculate
       the sampled gas volume at standard conditions.

-------
                 - 33 -


5.2.3.1  Chlorine
         An aliquot of each of the resulting impinger
         solutions was mixed with 0-tolidine.   The ab-
         sorption at 438 nanometers (nm) of yellow
         product was compared with those of standards
         prepared in the laboratory*
5.2.3.2  Cyanide
         An aliquot of the resulting solution was
         placed in a beaker and appropriately buffered.
         A cyanide ion-selective electrode and refer-
         ence cell were placed in the resulting solu-
         tion and the response detected by a laboratory
         millivolt meter was compared with that from
         standards of cyanide prepared in the labora-
         tory.

         Due to relative instability of cyanide sam-
         ples, analyses were conducted as soon as pos-
         sible after sample acquisition.
5.2.3.3  Total Phenolics
         Aliquot portions of the probe wash and filter
         fraction and the impinger catch fraction were
         kept refrigerated to prevent loss of phenol.
         In the laboratory, the pH of a 50-ml aliquot
         was adjusted to less than four with a 50-per-
         cent phosphoric acid reagent.  The phenol was
         steam distilled from a glass still.  After
         distilling, a 10-ml quantity of distilled
         water was added to the still and heating was
         resumed until an additional 10-ml volume of
         distillate was collected.  The total volume
         of the distillate was recorded and a 6-micro-
         liter portion was analyzed by gas chromato-
         graphy using a 5-percent Carbowax 20 M liquid
         phase on a Diatoport 80/100 S solid support
         in a 1/4-inch stainless steel column operated
         at 170°C and a hydrogen flame ionization de-
         tector.  The phetiol concentration in the dis-
         tillate was obtained in terms of parts per
         million in solution by referring the phenol
         peak area to those phenol standards prepared
         at the same time.

-------
                        - 34 -


       5.2.3.4  Total Sulfur Oxides

                An aliquot portion of the sodium hydroxide
                solution captured in the impingers and an
                aliquot of the solution from the aqueous
               -probe wash and corresponding glass-fiber
                filter were combined with sufficient barium
                chloride to result in the formation of barium
                sulfate with an excess of barium.  The tur-
                bidity of the resulting suspension of barium
                sulfate was measured using an ultraviolet
                spectrophotometer and the absorbance at 494
                nm compared with standard solutions prepared
                in the laboratory.


5.2.4  EPA Method 7 Sampling

       The sampling and analytical procedure for nitrogen
       oxides follows the procedure outlined in EPA Method
       7 (Appendix B).  Briefly, this method includes the
       evacuation of a 2-liter flask containing 25 ml of
       0.02-percent hydrogen peroxide — 0.03-percent sul-
       furic acid reagent.   The flask is subsequently filled
       with exhaust gas through a purged probe system.  The
       sampled gas volume is determined from the measurement
       of the volume, temperature,  and pressure of the sample
       flask.


       5.2.4.1  Nitrogen Oxides

                A small volume of dilute sulfuric acid and
                hydrogen peroxide absorbs the oxides of
                nitrogen upon standing.   In the laboratory,
                phenol disulfonic acid solution is added to
                the resulting solution and the absorbance of
                the resulting solution is measured at 420 nm
                and compared with the absorbance of known
                standards prepared  in the laboratory.


5.2.5  Grab Sampling

       Exhaust gas samples  were drawn by a vacuum pump through
       a probe and a short  (less than one foot) section of
       Tygon tubing to a gas sample container with stopcocks
       at the inlet and outlet of the container, and then
       through a  flowrate meter.  By minimizing the length
       of tubing,  the wall  effects  associated with this type
       of tubing  are considered minimal.   Exhaust gas tho-
       roughly flushed the  contents of the gas collection
       container  and tubing to ensure that air was completely
       displaced.   Subsequently, stopcocks at the container
       inlet and  outlet were closed.

-------
                 - 35 -
5.2.5.1  Acetylene
         An aliquot of the gas sample was injected
         into a gas chromatograph with a syringe.  A
         Carbowax column separated the acetylene which
         was subsequently measured with a flame ioni-
         zation detector operated at 90°C using 40 ml
         of helium carrier gas per minute.  The peak
         produced was identified and compared with
         those produced by known standards in gas mix-
         tures prepared in the laboratory.
5.2.5.2  Carbon Monoxide
         An aliquot of the gas sample was injected
         into the gas chromatograph with a syringe.  A
         molecular sieve 13X column separated the
         carbon monoxide which was subsequently meas-
         ured with a thermoconductivity detector main-
         tained at 50°C.  The peak produced was iden-
         tified and compared with those produced by
         known standards in gas mixtures prepared in
         the laboratory.
5.2.5.3  Ethylene and Homologues

         An aliquot of the gas sample was injected
         into a gas chromatograph with a syringe.  A
         Carbowax column separated the ethylene and
         homologues which were subsequently measured
         with a flame ionization detector operated at
         90°C using 40 ml of helium carrier gas per
         minute.   The peak produced was identified
         and compared with those produced by known
         standards in gas mixtures prepared in the
         laboratory.
5.2.5.4  Methane and Homologues

         An aliquot of the gas sample was injected
         into a gas chromatograph with a syringe.  A
         Carbowax column separated the methane and
         homologues which were subsequently measured
         with a flame ioni'zation detector operated at
         90°C using 40 ml of helium carrier gas per
         minute.  The peak produced was identified
         and compared with those produced by known
         standards in gas mixtures prepared in the
         laboratory.

-------
                        - 36 -
       5.2.5.5  Total Light Hydrocarbons

                An aliquot of the gas sample was injected into
                a gas chromatograph with a syringe.  A Carbo-
                wax column separated the total light hydro-
                carbons which were subsequently measured with
                a flame ionization detector operated at 90°C
                using 40 ml of helium carrier gas per minute.
                The peak produced was identified and compared
                with those produced by known standards of gas
                mixtures prepared in the laboratory.  Results
                were reported as total light hydrocarbons as
                methane.

5.2.6  Activated Charcoal Sampling

       The sampling train consisted of a stainless steel
       probe, polyvinyl chloride tubing, and a glass tube
       containing activated charcoal.  The tubing length
       used to transport the gases from the probe to the
       charcoal tube was kept to a minimum length — approxi-
       mately one foot.  This minimized wall effects as pos-
       sible sources of interferences.  The sampled exhaust
       gases were drawn at approximately one liter per minute
       through a calibrated dry-gas test meter, where the
       temperature and pressure of the gases were measured,
       and through a flowrate meter by means of a vacuum
       pump.  The tube containing the activated charcoal was
       capped for transport to the laboratory.

       The volume of metered gas was determined from the ini-
       tial and final readings of the gas meter.  The average
       meter pressure and temperature were used to calculate
       the sampled gas volume at standard conditions.

       Based upon Clayton Environmental Consultants' prior
       experience in the use of activated carbon for organic
       species sampling, the organic materials sample in
       this study using this method are considered to be quan-
       titatively absorbable and desorbable using carbon di-
       sulfide as the eluent.
       5.2.6.1   Benzene and  Homologues

                Benzene and  homologues  of benzene which were
                absorbed on  the  activated charcoal were de-
                sorbed  by stirring the  charcoal with carbon
                disulfide (1.5 ml  CS2  per gram charcoal) for
                a  minimum of 30  minutes.

                An aliquot of the  sample  was  injected into
                a  gas chromatograph containing a 5-percent
                Carbowax 20M TPA 2' x 1/4" column operated at

-------
                        - 37 -
                60°C.  The eluted fractions were measured by
                means of a flame ionization detector.  The
                peak areas produced by benzene and its homo-
                logues were measured and compared with those
                produced by known concentrations of the stand-
                ards prepared in the laboratory.
       5.2.6.2  Carbon Disulfide
                The carbon disulfide absorbed on the activated
                charcoal was desorbed by stirring the charcoal
                with ethanol (1.5 ml ethanol per gram charcoal)
                for a minimum of 30 minutes.

                An aliquot of the sample was injected into a
                gas chromatograph containing a 20-percent
                Carbowax 20M TPA 4' x 1/4" column operated at
                80°C.  The eluted fractions were measured by
                means of a thermoconductivity detector.  The
                peak areas produced by carbon disulfide were
                measured and compared with those produced by
                known concentrations of the standards in gas
                mixtures prepared in the laboratory.
       5.2.6.3  Pyridine
                The pyridine which was absorbed in the acti-
                vated charcoal was desorbed by stirring the
                charcoal with ethanol (1.5 ml of ethanol per
                gram of charcoal for a minimum of 30 minutes),

                An aliquot of the sample was injected into a
                gas chromatograph containing a 5-percent Car-
                bowax 20M TPA 21 x 1/4" column operated at
                60°C.  The eluted fractions were measured by
                the flame ionization detector.  The peak
                areas produced by pyridine were measured and
                compared with those produced by concentra-
                tions of standards prepared in the laboratory,
5.2.7  EPA Method 11 Sampling

       The sampling and analytical method for hydrogen sul-
       fide followed exactly the procedure of EPA Method 11
       (Appendix B).  Briefly, the method includes the cap-
       ture of hydrogen sulfide gas in a cadmium hydroxide
       solution in midget impingers immersed in an ice bath.
       The volume of sampled gas was measured with a dry-gas
       test meter.

-------
                        - 38 -
       5.2.7.1  Hydrogen Sulflde
                Analysis of the sample was conducted imme-
                diately after sample collection because of
                the instability of the resulting solution.
                A known quantity of iodine solution was
                added to the cadmium hydroxide solution and
                was subsequently back-titrated with a stand-
                ard solution of sodium thiosulfate.  The
                amount of hydrogen sulfide in the sample was
                calculated from the excess of iodine beyond
                that neutralized by the thiosulfate.
5.2.8  Sulfuric Acid Sampling

       The sampling train consisted of a sampling nozzle, a
       glass-lined probe, a heated 110-mm glass-fiber filter
       and two Greenburg-Smith impingers each containing 150
       milliliters of dilute sulfuric acid.   The impingers
       and a condensate trap containing silica gel were im-
       mersed in an ice bath.

       The sampled gas was drawn at approximately one CFM by
       means of a vacuum pump through a calibrated dry-gas
       meter where the temperature and pressure of the gas
       were measured, then through a calibrated orifice meter

       After each test the impinger contents and the rinsings
       of the impingers were transferred to  sample bottles
       for transport to the laboratory.  The volume of me-
       tered gas was determined from the initial and final
       readings of the gas meter.   The average meter pres-
       sure and temperature were used to calculate the sam-
       pled gas volume at standard conditions.
       5.2.8.1  Ammonia

                The total volume of the sample solution was
                measured and a 10-ml portion was transferred
                to a 50-ml Griffin beaker to which a 4-ral
                quantity of 0.2-molar sodium citrate reagent
                was added.  The pH of the solution was ad-
                justed to 7.5 using dilute acid and base and
                an expanded scale pH meter.   The final volume
                was adjusted to 20-ml with 0.2-molar sodium
                citrate.  The ammonium ion potential of the
                solution was measured using  the millivolt
                mode of a Sargent Model DR pH meter and an
                Orion ammonia ion-selective  electrode, used
                in conjunction with an  Orion single junc-
                tion reference electrode.  The ammonia con-
                tent (as NH-j) of the sample  was calculated

-------
                        - 39 -
                by reference to a standard curve prepared by
                application of the procedure to a series of
                samples of known ammonium content.
5.2.9  Ethanol Sampling

       The sampling train consisted of a sampling nozzle, a
       glass-lined probe, and two midget impingers contain-
       ing 30 milliliters of ethanol.   The impingers and a
       condensate trap were placed in  an ice bath.  The sam-
       pled gas was drawn at approximately one CFM by means
       of a vacuum pump through a calibrated dry-gas meter
       where the temperature and pressure of the gas were
       measured.

       After each test the impinger contents were transferred
       to sample bottles for transport to the laboratory.
       The volume of metered gas was determined from the ini-
       tial and final readings of the  gas meter.
       5.2.9.1  Acrolein

                An aliquot of the resulting solution was mixed
                with 4-hexylresorcinol in an ethyl alcohol
                trichloroacetic acid solvent in the presence
                of mercuric chloride.   The absorption at 605
                nm of the blue product was compared with
                standards prepared in the laboratory.
5.2.10 Cyclohexane Sampling

       The sampling train consisted of a sampling nozzle, a
       glass-lined probe, a heated 110-mm glass-fiber filter,
       and two Greenburg-Smith impingers each containing 150
       milliliters of cyclohexane.  The impingers and a con-
       densate trap containing silica gel were immersed in
       an ice bath.

       The sampled gas was drawn at approximately one CFM by
       means of a vacuum pump through a calibrated dry-gas
       meter where the temperature and pressure of the gas
       were measured, then through a calibrated orifice meter

       Similarly, the probe was rinsed with cyclohexane and
       the filter was added to the rinsings.   "Filterable"
       material was captured in this sample fraction.

       After each test, the impinger contents and the rins-
       ings of the impingers were transferred to sample bot-
       tles for transport to the laboratory.   The volume of
       metered gas was determined from the initial and final

-------
                 - 40 -
                                        ;|
readings of the gas meter.  The average meter pres-
sure and temperature were used to calculate the sam-
pled gas volume at standard conditions.

To minimize the effects of interferring substances
having the same retention time in the gas chromato-
graph as the compound of interest thus biasing the
results obtained from this type of sampling, exact-
ing standards-preparation procedures and blank-cor-
rection procedures were designed and carefully
executed.  For each compound of interest (delineated
below), the chromatogram peak areas were adjusted
according to standards and according to replicate
analysis performed on the pure solvents and on field
blanks.  Further, Clayton's laboratory experience
with similar coke-oven studies provided expertise
helpful in recognizing and minimizing interference
problems in the analytical techniques.
5.2.10.1 Acridine

         An aliquot of each solution was injected into
         a gas chromatograph containing a 6-percent,
         Dexsil 300 on an 80/100 Chromosorb W column,
         in an oven temperature programmed from 165°C
         to 300°C at 4°C per minute.  A flame ioniza-
         tion detector indicated acridine in the sam-
         ple.  The peak was compared with those pro-
         duced by known standards prepared in the
         laboratory.
5.2.10.2 Anthracene

         An aliquot of each solution was injected into
         a  gas  chromatograph containing aj 6-percent,
         Dexsil 300 on an 80/100 Chromoso'rb W column
         operated at 165°C to 300°C at 4°C per minute.
         A flame ionization detector indicated the
         presence of anthracene in the sample.  The
         peak was compared with those produced by
         known standards prepared in the laboratory.
5.2.10.3 Benzo(a+e)pyrene

         An aliquot of each solution was injected into
         a gas chromatograph containing a 6-percent,
         Dexsil 300 on an 80/100 Chromosorb W column
         operated at 165°C to 300°C at 4°C per minute.
         A flame ionization detector indicated the
         benzo(a+e)pyrene in the sample.  The peak
         was compared with those produced by known

-------
                 - 41 -
         standards in gas mixtures prepared in the
         laboratory.
5.2.10.4 Carbazole
         An aliquot of each solution was injected into
         a gas chromatograph containing a 6-percent,
         Dexsil 300 on an 80/100 Chromosorb W column
         operated at 165°C to 300°C at 4°C per minute.
         A flame ionization detector indicated the
         carbazole in the sample.   The peak was com-
         pared with those produced by known standards
         in gas mixtures prepared  in the laboratory.
5.2.10.5 Chrysene
         An aliquot of each solution was injected into
         a gas chromatograph containing a 6-percent,
         Dexsil 300 on an 80/100 Chromosorb W co umn
         operated at 165°C to 300°C at 4°C per minute.
         A flame ionization detector indicated the
         presence of chrysene in the sample.  The peak
         was compared with those produced by known
         standards prepared in the laboratory.
5.2.10.6 Fluoranthene
         An aliquot of each solution was injected into
         a gas chromatograph containing a 6-percent,
         Dexsil 300 on an 80/100 Chromosorb W column
         operated at 165°C to 300°C at 4°C per minute.
         A flame ionization detector indicated the
         presence of fluoranthene in the sample.   The
         peak was compared with those produced by
         known standards prepared in the laboratory.
5.2.10.7 Fluorene
         An aliquot of each solution was injected into
         a  gas  chromatograph  containing  a 6-percent,
         Dexsil 300 on an 80/100 Chromosorb W column
         operated at 165°C to 300°C at 4°C per minute.
         A  flame ionization detector indicated the
         presence of fluorene in the sample.  The peak
         was compared with those produced by known
         standards prepared in the laboratory.

-------
                         - 42  -


       5.2.10.8  9-Fluorenone
                 An aliquot of each solution was injected into
                 a gas chromatograph containing a 6-percent,
                 Dexsil 300 on an 80/100 Chromosorb W column
                 operated at 165°C to 300°C at 4°C per minute.
                 A flame ionization detector indicated the
                 presence of 9-fluorenone in the sample.  The
                 peak was compared with those prpduced by
                 known standards prepared in the laboratory.
       5.2.10.9  Naphthalene
                 An aliquot of each solution was injected into
                 a gas chromatograph containing a 6-percent,
                 Dexsil 300 on an 80/100 Chromosorb W column
                 operated at 165°C to 300°C at 4°C per minute.
                 A flame ionization detector indicated the
                 presence of naphthalene in the sample.  The
                 peak was compared with those produced by
                 known standards prepared in the laboratory.
       5.2.10.10 Phenanthrene
                 An aliquot of each solution was injected into
                 a gas chromatograph containing a 6-percent,
                 Dexsil 300 on an 80/100 Chromosorb W column
                 operated at 165°C to 300°C at 4°C per minute.
                 A flame ionization detector indicated the
                 presence of phenanthrene in the sample.  The
                 peak was compared with those produced by
                 known standards prepared in the laboratory.
       5.2.10.11 Pyrene
                 An aliquot of each solution was injected into
                 a gas chromatograph containing a 6-percent,
                 Dexsil 300 on an 80/100 Chromosorb W column
                 operated at 165°C to 300°C at 4°C per minute.
                 A flame ionization detector indicated the
                 presence of pyrene in the sample.  The peak
                 was compared with those produced by known
                 standards prepared in the laboratory.
5.2.11 Integrated Sampling

       Exhaust gas samples were drawn by a diaphragm pump
       through a probe and tubing to a Tedlarw bag.  A
       valve at the inlet to the bag was then closed.  The
       gas sample in the bag was transported to the labora-
       tory for analysis.

-------
                        - 43 -


       5.2.11.1  Total Reduced Sulfur

                 An aliquot of the gas sample was introduced
                 into a gas chromatograph.  A 3-percent
                 OV-1, 80/100 mesh Chromosorb separation
                 column (6' x 1/4") isolated the total reduced
                 sulfur compounds which were subsequently
                 measured with a flame photometric detector.
                 The peaks produced were identified and com-
                 pared with those produced by known standards
                 in gas mixtures prepared in the laboratory
                 of Ontario Research Foundation, a subcon-
                 tractor to Clayton.

                 Due to the tendency of reduced sulfur com-
                 pounds to degrade rapidly, the gas samples
                 were analyzed as soon as possible after sam-
                 ple acquisition.  Because the analyses for
                 total reduced sulfur require non-portable
                 analytical instrumentation, the samples were
                 transported to the laboratory and analyzed
                 within 24 hours.


5.2.12 Andersen Impactor Sampling

       The size distributions of particles at the inlet and
       outlet sampling locations were determined by using an
       Andersen impactor.   Exhaust gases were drawn from the
       stack gas and passed through a series of impaction
       nozzles and plates which collect particles of pro-
       gressively smaller aerodynamic size.  Those particles
       not impacted on the final stage were captured on a
       glass-fiber filter located behind the impaction
       separator.

       During the sampling for particulate size distribution,
       the flowrate through the impactor was chosen so as to
       provide as near as possible an approximation to iso-
       kineticity at the single point of sampling in the duct
       cross-section.  The expected velocity pressure at the
       sampling point was determined from velocity pressure
       measurements made in previous tests for particulate
       emissions.  Once set, that flowrate through the im-
       pactor was maintained throughout the push duration.


       5.2.12.1  Particle Sizing

                 The weight of material captured on each stage
                 was subsequently determined in the laboratory
                 on an analytical balance.  The final filter
                 was also weighed in the laboratory on an
                 analytical balance.  The relative quantity

-------
                        - 44 -


                 of weight captured on the various impaction
                 stages and the final filter indicates the
                 size distribution of collected particulate
                 materials.  Subsequently, the density of
                 particulate was determined in the laboratory
                 by a pycnometer in order to establish the
                 "cut-off" diameter for the various impactor
                 stages.

                 The operating instructions for the Andersen
                 impactor prescribe that certain correction
                 factors  be applied to the size distribution
                 results  and cut-off diameter for each stage.
                 The unit has been calibrated by the supplier
                 for spherical, unit-density particles.  Sepa-
                 ration,  based on the particle's inertia, from
                 stage to stage depends upon the particle's
                 aerodynamic diameter.  Charts provided by
                 the supplier permit calculation of actual
                 cut-off  diameter for each stage for the par-
                 ticular  material being sampled and the actual
                 stack gas sampling conditions.  The multipli-
                 cative correction factor for particle density
                 requires that a density determination be made
                 which is not a bulk density type measurement
                 but, rather, it must estimate, as closely as
                 possible, the density of the discrete mate-
                 rial as  it existed under stack gas conditions.
5.2.13 Fugitive Emissions Measurement (Method 1)

       In an attempt to corroborate estimates of  fugitive
       particulate emissions made by the visual judgment
       method described in Section 5.1.3 (Method  2),  a sam-
       pling technique was developed using a combination of
       photography and particulate concentration  measurements
       in the plume of the fugitive emissions.   This  technique
       is termed "Method 1."  Although well-defined,  widely-
       accepted procedures exist  for the measurement  of emis-
       sions which are confined  in a duct or stack,  standard-
       ized procedures for the measurement of fugitive emis-
       sions are non-existent.  Access to the collection hood
       and the potentially-hazardous environment  near the
       hood during coke pushing  are two factors that  constrain
       the scope of the technique to one of simplification and
       relative ease of implementation.

       Fugitive emissions were thus estimated by  two  separate
       techniques in order to assess their contribution to
       total pushing emissions:   1) consideration of  the rela-
       tive fugitive plume concentration both when emissions
       were completely uncontrolled and when the  emission con-
       trol system was in operation; and 2) subjective evalua-
       tion by an engineer-observer.  This section describes
       the first of these two techniques.

-------
                 - 45 -
5.2.13.1  High-Volume Sampling and Analytical Consid-
          erations

          One high-volume sampler was positioned above
          the hood as shown in Figure 5.2.13.1.  The
          axis of the high-volume sampler was placed
          nearly horizontally over the front of the
          hood car; the distance from the top of the
          hood to the sampler was approximately 10
          feet.  This zone, being close to the source,
          is highly turbulent immediately after push-
          ing occurs.  This turbulence is due largely
          to vortex flow caused by the high heat flux
          of the coke-push plume.  In addition to
          vortex flow, convective turbulence caused by
          cold air entrainment is taking place.  By
          observation, this turbulent area likely
          causes the push emissions to be well-distri-
          buted and well-mixed in this region.  Iso-
          kinetic sampling is impossible to attain in
          such a flow situation.  Because of hood
          movement from oven to oven, use of additional
          high-volume samplers poses serious operational
          difficulties.

          By placing the high-volume air sampler in a
          position that is invariant relative to the
          hood, the fugitive emissions measurement
          technique yields ratios of reproducible
          (precise and not necessarily accurate) fume
          concentrations above the hood (see Section
          5.2.13.2).  These ratios provide a measure
          of the hood capture efficiency as it varies
          with the greenness of the push.   In such a
          location, the time-integrated concentration
          provides a measure of the plume's concentra-
          tion throughout a push of given greenness at
          the point where the sampler is placed.  While
          somewhat wide fluctuations in concentration
          were found for samples acquired from a set
          of pushes having a given greenness, averag-
          ing by the use of linear regression (as dis-
          tinguished from arithmetic averaging over
          all greennesses by "lumping" all data) over
          the range of push greenness provides a good
          estimate of an index of relative plume con-
          centration.  This estimate can then be used
          for subsequent approximations of hood cap-
          ture efficiencies.

          Once the fugitive particulate was collected
          by the sampler, the filter incremental weight
          and total volume of air sampled (determined
          from a calibrated flowrate — see Appendix D —
          over a measured time period) were used to

-------
                 - 46 -
          calculate the average concentration of par-
          ticulate in the fugitive emissions plume
          which passed the hood at a representative
          location.  Temperature corrections were not
          made in calculating relative concentrations
          under the assumption that the absolute tem-
          perature did not vary appreciably in the
          turbulent, well-mixed push plume.  The photo-
          graphs of each sampled push provided a qual-
          itative characterization of the push plume
          to further ensure that only "similar" plumes
          were compared in the statistical analyses.
          The push plume was photographed such that a
          perspective view was available of fugitive
          emissions from the hood perimeter from two
          vantage points at right angles to one
          another.  These pictures also clearly dis-
          play the magnitude of the emissions plume.

          In the laboratory, the high-volume filter
          was dried at 65°C and weighed to determine
          the amount of filterable particulate col-
          lected during the fugitive emissions sam-
          pling.  Examination of the filters indi-
          cated that captured particulate matter was
          highly embedded in the filter matrix; good
          particle adhesion was evident.
5.2.13.2  Theoretical Considerations

          In any measurement of mass emission rates,
          it is necessary to extract a sample of the
          particulate-laden gas which is representa-
          tive of the time-space average concentra-
          tion of particulate in the gas passing a
          representative flux area.  A differentially
          small volume of gas containing a substance
          of concentration"C(A)" flows through a dif-
          ferentially small, normal area "dA" at a
          local velocity of "U(A),M a function of
          position;  integrating over the entire area
          of flow, "A," provides a spatial average
          estimate of the total mass flowrate of the
          substance  contained in the gas.   The follow-
          ing vector integral determines the mass flux
          through the total area:
          dM
          dt
ll   C(A)    [lj(A)  •  dt]             (1)

-------
       - 47 -
where:

       dM
       -rr = mass flux rate, Ibs/hr


     C(A) = local concentration of particu-
            late in the elemental area, dA

U(A) • dA = the vector scalar product of the
            local flux velocity vector with
            the normal component of the ele-
            mental area, dA.

In practice, average gas velocity and con-
centration must be approximated, particularly
in a harsh environment where local measure-
ments are impractical or impossible.  Fur-
ther, the flux area, through which mass is
transferred, is difficult to measure when
the emissions are fugitive in nature and
consequently not confined to a duct of known
area but rather pass unrestrained into the
air.  Hence, the following approximation is
often used:


4| 2: C(A)  [~U(A) • T] ,                 (2)

where the "bar" denotes an average value of
the variable and the symbol "=" denotes an
approximation.  Over a "large number of
similar pushes," (i.e., pushes of the same
greenness rating and plume behavior) the
bracketed term is approximately constant.
Under such an assumption, the mass emission
rate is proportional to the average concen-
tration in the plume.  If subscripts "1" and
"2" denote the off and on condition of the
exhaust fan, respectively, then equation (2)
can be used to write the following:

dMi/dt   GI
dM2/dt ~ Cl"                             ^3^

Furthermore, for a given push greenness rat-
ing, the ratio of average concentrations in
the entire plume may be replaced by a ratio
of "representative" concentrations which re-
flect the percentage change in concentration
at a fixed location as the hood fan is turned
on or off.  Now the following two mass balances
can be written:

-------
        -  48  -
fjpl
 dt
      ~d
      ~d
                  (fan  off)
       dF2    dI2
"dF" =  TT  +  ~dT   (fan  on)

where:

-r—  = source  emission rate,  Ibs/hr
                                         (4)


                                         (5)
j t?
-r- =  fugitive  emission  rate,  Ibs/hr
at
dl
•jf = captured emission  rate,  Ibs/hr
For pushes of  the  same  greenness  rating,  the
source emission  rate  is  somewhat  constant,
Therefore, the approximation  is made  that:
and, with the hood exhaust  fan  off,  the  cap-
tured emission rate  is  essentially  zero,

ail
TT - ° •

Therefore, the equations  combine  to  give:
dFi
_ ~_
 dt
dF2
_ ".
 dt
            dlo
            _ ~.
             dt
Equation  (8) may be normalized  through  divi-
sion by dF2/dt.  After  rearranging,
dI2/dt _ dF
dF2/dt ~ dF2/dt
Using the approximation set  forth  in  equation
(3), equation  (9) is rewritten,

dI2/dt   Ci
Now, source sampling in  the duct which  ex-
hausts the hood provides an integrated  sam-
ple of the collected emissions  captured  by
the hood during the push whose  duration  is
"T."  Denoting the measure of these  in-duct
emissions, as computed from the source  sam-
pling data, by R,

-------
        -  49  -
       »T

*  =   I  I T^)  "  .                     (11)
      'o
Then,  equation  (10)  can  be  rewritten,

                                         (12)

 If  GI  and  €2  are measured  averages  over  time
 (during  the entire  push),  and  are measured
 at  representative,  well-mixed  locations  above
 the collecting hood,  then  the  integral  sim-
 plifies  to:
  £l  -  1 ]   E  &
  ,Co    /
(13)
where:

E  =  I  (-jjj-)  dt                         (14)

and  E represents  an  estimate  of  the  actual
mass emission  rate of  fugitive emissions
around the hood,  emitted  over the  entire
push, with the hood  exhaust  fan  on.   Re-
writing equation  (13),

»  .       R
An estimate of  the  absolute  fugitive  mass
emission  rate from  the  collection  hood  can
be calculated from  the  measured  particulate
emission  rate in  the duct which  exhausts
the hood  and  from the ratio  of representa-
tive,  time-average  particulate concentra-
tions  above the hood measured during  similar
•pushes with the exhaust fan  off  and on.

Equation  (15) provides  a relatively good
estimate  of the fugitive emissions to the
extent that the following assumptions are
valid :

      1.   Plume  behavior of  fugitive emissions
          around and above the hood is similar
          for  similar process conditions  (i.e.,
          same push  greenness).

      2.   The  air  sampler is  located at  a
          location above the  hood such that
          the  concentration measured by  the

-------
        -  50  -
         sampler is proportional  to  the
         average plume concentration at  that
         elevation, irrespective  of  whether
         the exhaust  fan is off or on.

     3.  The product  of the plume cross-sec-
         tional area  and rise velocity
         through the  planar sampling area
         is somewhat  constant for similar
         pushes and similar meteorological
         conditions.

To characterize the weather conditions dur-
ing the fugitive emissions sampling  period,
as compared with "normal conditions," mete-
orologists from Clayton Environmental Con-
sultants studied local climatological records,
The data were recorded at Detroit Metropoli-
tan Airport located about 20 miles southwest
of the Rouge Plant.   Such data are useful in
showing that the daily average wind  speed and
direction observed on test days were  repre-
sentative and not atypical when compared with
yearly averages.

The meteorological conditions recorded dur-
ing the testing at the Ford Motor Company
complex in Dearborn, Michigan, were  near
normal based upon current and historical
climatological data.  The tests performed
on two days in January occurred when wind
speeds were above average, with one  day
having above-average temperatures.   The
tests performed in February consisted of
two days with above-average wind speeds, and
one day of well-below-average wind speeds.
Temperatures were normal during these dates
as were the wind directions.   All tests con-
ducted in April were on days with below-
average wind speeds and above-average tem-
peratures.   For seven days of testing during
June, wind speeds were below average.  How-
ever, for the entire month, the combined
average wind speed was near normal.   Tem-
peratures were also near normal during this
month, as were the wind directions.   Average
wind speeds for the testing dates during the
month of July were slightly below normal.
Average(temperatures,  however, were very
close to normal, as were the wind directions.

-------
                 - 51 -
5.2.13.3  Qualifications
          We emphasize that the use of high-volume air
          sampling to characterize fugitive emissions
          is a largely unproven technique.   It is rea-
          sonable to expect variation in the flow terms
          (i.e.,  bracketed term in equation (2)) from
          push to push even under conditions of "con-
          stant"  push greenness and uniform meteoro-
          logical conditions due to behavioral differ-
          ences in plume formation and rise.  The ex-
          tent to which these differences materially
          effect  the approximation made to  derive
          equation (3) is largely unknown and  not
          documented in this study.  Therefore, it is
          emphasized that the derived fugitive mass
          emission rates and hood capture efficiencies
          used throughout this report are qualified in
          this respect.  Accordingly, all estimates
          of total system abatement efficiency and all
          overall system emission factors are  limited
          with respect to accuracy to the extent that
          the two estimating techniques for fugitive
          emissions do not approximate "true"  hood
          capture efficiency.

          Although Clayton Environmental Consultants
          believes the fugitive emissions estimates
          derived from these methods reasonably appro-
          ximate  actual fugitive emissions  and add
          valuable information concerning emissions
          from this control system, we recognize that
          these techniques are unproven and rest on
          certain unverified assumptions.

-------
                             - 52 -


            6.0  PRESENTATION AND ANALYSIS OF RESULTS
To draw meaningful conclusions from the plethora of data collected
during this study, techniques were developed to organize, compile,
correlate, and statistically analyze the available information.
Methods of compilation and presentation of the data are necessar-
ily diverse due to the variety in quantity and quality of his-
torical data, operating and economic data, process data, and per-
formance data for both the hood capturing device and the scrubber.


6.1  Organization and Compilation

     6.1.1  Historical Data

            The data contained in the daily operating checks are
            used to characterize system performance in two ways:
            1) to determine whether the control system perform-
            ance during the source testing period was typical and
            representative of historical performance, and 2) to
            determine utility costs throughout the history of the
            control system.  Consequently, the data contained on
            the daily operating checks were used to compute
            monthly averages for each of these variables through-
            out the history of the control system.

            Table 6.1.1-1 displays the variation in fan current,
            water flowrate to the Kinpactor, and calculated air
            and water horsepower required during the period
            August, 1973 through August, 1975.  This 2-year sum-
            mary shows that considerable fluctuations occur in
            the monthly averages for these quantities.  These
            data are useful in validating test results within the
            context of previous system performance.

            The criteria used to judge the representativeness of
            the process conditions during testing was that the
            average value of an operating parameter during test-
            ing should fall within an interval of +10% of the
            long-term average value of that variable.(6)   Table
            6.1.1-2 also shows that typical process conditions
            existed with respect to the scrubber sampling.  Fan
            currents between and during coke pushes were +9.4%
            and +7.2% above the historical measurements, respec-
            tively.  Kinpactor throat differential pressures be-
            tween and during coke pushes were +10% and +2.5% above
            historical record, respectively.  Water flowrate to
            the scrubber and pump pressure were -13% and -7.6% of
            the historical record, respectively.

            To determine the utility consumption throughout the
            history of the control system and to monitor how this
            consumption varies with time, the major items contri-
            buting to utility costs were determined on an average

-------
                 - 53 -
monthly basis from the historical data.   Fan  HP  require-
ments were determined by electrical consumption  during
both the push cycle and idle cycle for the  control  sys-
tem.  Since the motor operates at 4160 volts, multipli-
cation of this voltage by fan current read  from  the
strip chart recorder determines electrical  consumption.
(Additionally, because the motor is three-phase,  the
factor 1.73 was applied; the power factor was assumed
to be 1.0.)

Water pump power consumption was computed from flowrate
data as monitored on the flow recorder in the control
room and from pump pressure as measured by  gauges on .the
pump outlet.  The equation used in computing  power  re-
quirements for the water pump is as shown:


HP  =   Pump pressure (ft. water) x flowrate  (gpm)
                       3960 x (0.85)
where a pump efficiency of 85 percent has been assumed.

Table 6.1.1-1 summarizes the utility requirements com-
puted by these methods for the major utility consumers -
the induced draft fan and the water pumps.  ,By apply-
ing weighting factors determined by prolonged observa-
tion of the control system and typical operator tenden-
cies at Ford Motor Company, the total utility require-
ment over the two-year period is computed.  By assuming
that the idle cycle of the control system comprises
approximately 40 minutes of any one push-hour while the
push cycle (when the emission control system is energized
by placing the coke guide in alignment with the coke
oven to be pushed) takes approximately 20 minutes of any
push-hour, the total utility requirements are shown
in Table 6.1.1-1.

Figure 6.1.1 presents a graphical display of the year-
to-date cost information obtained from cost accounting
departments.  The costs, plotted separately for 1974
and 1975, are useful in indicating whether there is any
tendency toward excessive maintenance due to deteriora-
tion of the control system from year-to-year.  Because
the cost data are necessarily limited by cost accounting
procedures', this method of presentation and analysis
of historical economic data was the only.technique that
could be devised for this study.

The data contained in maintenance or operational logs
are difficult to characterize and display.  Except for
certain recurring problems, many of the malfunctions
documented in these logbooks must be read in their

-------
                        - 54 -
       entirety to appreciate the nature of the downtime caused
       by the malfunction.   However,  in an attempt to extract
       salient information  from the logbook entries,  a summary
       table has been constructed which presents the  major causes
       of system downtime.   Table 6.1.1-3 displays these data.
       Under each major source of system downtime are listed
       the dates during which some downtime was attributed to
       that particular type of system malfunction.  Conclusions
       can be drawn from such a summary as to either  the inter-
       mittent or ongoing failure of  certain system components.
       In addition to this  type of presentation, direct reading
       of logbook entries pertaining  to the "Smokeless Coke
       Pushing System" maintenance provides a good overview of
       the type of malfunctions and the importance of the mal-
       functions.  All entries are presented in Appendix A.


6.1.2  Process and Engineering Data

       Appendices G and H,  respectively, include the  "Daily Log
       of Fume Hood Performance" and  "Scrubber Performance" data
       sheets (to the extent that they were completed by the
       on-site engineer-observer).  These data, in total, are
       used to determine overall system performance as it varies
       throughout time.  For those days on which source testing
       occurred, the data from the engineer-observer's log
       records are assembled into special tables which summarize
       the process conditions during  each type of test conducted.

       Appendix I presents  tables that list the pertinent pro-
       cess parameters and  their averages as measured during the
       source tests.  These tables summarize the oven numbers,
       greenness of the. push, cross-draft wind velocity, stack
       gas opacity, and net coal charged to each oven pushed
       during the course of any given test.  The averages of
       all such pushes included in the test under consideration
       are also included.  These test averages are used as the
       basic engineering and process  data for subsequent correla-
       tions presented in Section 7.0.

       Also included in these tables  are daily averages of the
       scrubber inlet and outlet dry-bulb and wet-bulb tempera-
       tures, scrubber inlet and outlet static pressures, throat
       pressure drop, water pressure  at the scrubber  throat,
       recirculated water flow, and fan motor current.  As previ-
       ously indicated, these latter  quantities vary  insignifi-
       cantly through the course of a day's testing.   As such,
       the daily averages are reported as they are obtained from
       any readings taken by the engineer-observer while he was
       not participating in hood capture efficiency estimates
       during a source test.

       These tables would have also included net coking time;
       however, due to a malfunction  in the Ford Motor Company

-------
                             - 55 -
            recording device, the time of coal charge for each
            particular oven was not available.  As such, only gross
            coking time estimates could be made from the times in-
            cluded on the coal charge log sheets, and these are
            included on the process data summaries.


     6.1.3  Scrubber Emissions Data

            Tables 6.1.3-1 through 6.1.3-92 present mass emission
            rates and concentrations for each contaminant measured
            at the scrubber inlet and/or outlet.  These data and
            the process data presented in Section 6.1.2 are related
            and presented in Section 7.0.


     6.1.4  Fugitive Emissions Data

            Tables 6.1.4-1 to 6.1.4-5 present the data used in deter-
            mining filterable particulate concentrations above the
            hood.  Three different system configurations are included
            in the data base:  1) no collection hood present above
            the pushed oven; 2) hood in place but no exhaust from
            scrubber fan (i.e., fan louvers closed); and 3) hood in
            place and operating with full exhaust volume.  The last
            two configurations provide the data base used in subse-
            quent correlations (see Section 6.3) for providing the
            quantitative measure of hood collection efficiency de-
            fined in Section 5.2.13. To assist in relating these data
            to any simultaneous source sampling measurements and to
            the process data, the dates, push times, and oven numbers
            are provided.  A line of data marked with an asterisk
            indicates a data point classed as an outlier (see Section
            6.3 for methodology).


6.2  Emission and Performance Models

     6.2.1  Push Emissions

            In order to normalize emissions to process parameters,
            it is necessary to estimate the emission rate of all
            coke-pushing emissions whether they emanate from the
            scrubber stack outlet or from the perimeter of the col-
            lection hood.  Since relative emission rates of fugitive
            emissions were measured on a concentration basis, it is
            necessary to consider mass balances of particulate emitted
            during a push in an effort to estimate total system abate-
            ment efficiency.  Figure 6.2.1 schematically depicts all
            streams which contain pollutants entering and exiting
            a hypothetical boundary drawn around the control system.
            Further, a mass balance computed by drawing a hypotheti-
            cal boundary and considering all streams into and out of
            the collection hood permits relation of fugitive emissions

-------
                 - 56 -
to scrubber inlet emissions.  Recalling that the only
measurements of mass emission rates made directly were at
the scrubber outlet and scrubber inlet, it is necessary
to compute fugitive mass emission rates using the follow-
ing mass balances (which do not explicitly account for
door leaks).
     P

     P
   F + 0 + S

   F + I
(1)

(2)
where,

     P


     S

     F

     I

     0
=  total pushing emissions from oven,
   Ibs/hr

=  emissions removed by scrubber, Ibs/hr

=  fugitive emissions from hood,.Ibs/hr

=  emissions captured by hood, Ibs/hr

-  scrubber outlet emissions, Ibs/hr
Now, since hood capture efficiency is defined as:
     E
      H
   1 - F/P
(3)
if terms are regrouped and it is recognized that,

     F  •  P - I                                (4)

then,

     EH •  I/P.                                 (5)

Now, having measured the inlet emission rate to the scrub-
ber (I), the fugitive emission rates from the hood can
be computed on a mass emission rate basis if the hood
capture efficiency has been estimated by independent meas-
urements.

As indicated in Section 5.2.13, concentration measurements
around the hood perimeter were made to estimate hood
capture efficiency.  These measurements resulted in a
linear regression line determined and fitted to the data
which provided an estimate of hood capture efficiency as
a function of the degree of coke greenness discussed be-
low.  Furthermore, employing the same data base, linear
regression lines may be computed considering the engineer-
observer's visual estimate of hood capture efficiency as

-------
                 - 57 -
a function of the degree-of-greenness.  These  two
regression lines, the methodologies of which are discussed
in Section 6.3, are used in establishing a range of  esti-
mate of hood capture efficiency.  When this range  is
used in conjunction with the equation shown above, a range
of fugitive mass emission rates can be computed.   Since
the scrubber efficiency is defined as
     E0  =  1 - 0/1                              (6)
      s
and total efficiency is defined as
     ET  =  1 - _°+IL,   ,                        (7)
substitution yields



         E~  T? TT                                  ( Q \
      T  ~  ^TT^O                                 \° )
Thus, the total system efficiency is the product of the
hood capture efficiency and the scrubber control effi-
ciency.  This definition of total efficiency is reason-
able in that system abatement is as much dependent upon
efficient capture of pushing emissions as it is upon
efficient scrubbing.  Either efficiency, when low, results
in a low system efficiency.

Equation (8) shows clearly the st/bng influence that the
hood capture efficiency, EJJJ plays in determining total
system abatement efficiency, EX.  For example, in con-
sideration of the relatively high scrubber efficiencies
in abating particulate matter, a 50-percent hood capture
efficiency translates to an approximate 50-percent
total system abatement efficiency.  Accordingly, any
error in the estimate for hood capture efficiency (see
Section 5.2.13)introduces a proportional error in total
system abatement efficiency.  Therefore, estimates of
total system abatement efficiencies are only as accurate
as the range of estimates for hood capture efficiencies.

Although the measurements conducted to determine the re-
gression lines for estimating hood capture efficiency
were determined for particulate matter, it is assumed in
subsequent calculations that the hood capture efficien-
cies can be applied equivalently to the gaseous contami-
nants .

The highly turbulent flow patterns above the hood, coupled
with relatively low indraft velocities present during push-
ing at the hood perimeter, render the assumption relative-
ly accurate.  The assumption permits calculation of total

-------
                        - 58 -
       abatement efficiency (coupling fugitive emission losses
       and scrubber losses) for gaseous contaminant removal.
       This "model" is therefore used implicitly in subsequent
       tables presented in Section 7.0 which display total
       system efficiencies for gaseous and particulate con-
       taminants .
6.2.2  Scrubber Performance

       While many mechanisms and models have been postulated
       to account for scrubber efficiency for both particulate
       and gaseous contaminants, correlation of scrubber effi-
       ciency to "contact power" remains one of the most straight-
       forward correlations available.

       Theoretical attempts to explain variations in, or even
       the absolute value of, scrubber efficiency are usually
       successful and accurate if large amounts of empirical
       data are available for highly random processes.  Correla-
       tion of efficiency data is attempted here to lend credi-
       bility to the test results and to extrapolate the per-
       formance of the system to other conditions.

       In contact power correlations, '  '  the penetration (or
       one minus the efficiency (l-Eg)) is defined in terms of
       the number of transfer units, Nt.  The relationship is
       logarithmic as shown in the following equation:


            N   =  In (	\	^ .                     (1)

                      V l ~ E*J
       On a log-log plot of the number of transfer units as a
       function of contact power, efficiency data can be repre-
       sented as a straight line having a slope and intercept
       which is a function of the particular scrubber geometry
       and contaminant being removed.  The contact power is
       computed for each set of scrubber conditions by summing
       the contact power put into the scrubber by way of the
       gas stream and the liquid stream.  Contact power derived
       from the gas stream is given by:


            Pg  =  0.157 (Ap).                        (2)

       The pressure drop, Ap, used in computing the gas con-
       tact power is measured across the unit at points where
       gas velocities are equal.  The pressure drop inherently
       must be a measure of useful work done in the scrubbing
       process and cannot include pressure changes due to ve-
       locity in the gas stream.  The contact power derived from
       the liquid hydraulic nozzles is calculated from:


            PL  =  0.583 (Pp) £m                   (3)

-------
                 - 59 -
where:
     P   =  pressure at the nozzles (psig)
  qL/q
G
      liquid to gas ratio (gpm/1000 SCFM).
The equation assumes that the residual kinetic energy
of the liquid stream after contacting with the gas
stream is negligible.  This contact power, it is assumed,
has totally dissipated into turbulent mechanisms which
efficiently extract particles from the gas stream.

Table 6.1.1-2 lists the monthly averages of contact
power over the two-year history of the control system,
as computed from the data contained in Table 6.1.1-1.
While the contact power varies considerably over the
two-year history of the system, the monthly average con-
tact powers during the source testing period (June and
July, 1975) are relatively close  to the overall aver-
ages.  Therefore, we conclude that efficiencies deter-
mined during testing were typical of historical scrub-
ber performance as it can be established throughout
the system lifetime.

The utility of contact power correlation rests in a
proper set of data which permits a good line to be
drawn in a contact power graphical plot.  It can be
seen from the contact power summaries and the emission
data summaries, that the range of values of contact
power encountered during the source sampling period of
June and July, 1975 are not sufficient to provide a
broad enough range of contact power for a good correla-
tion.  As such, it is not possible to extrapolate with
any degree of accuracy what the performance of the
scrubber would be if it was operating  in accordance
with the design manual specification  — a pressure drop
of 60 inches of water.  While it is possible to assume
a slope of the line in the contact power correlation
and pass that line through the available data point,
the slope of the line is a sensitive function of par-
ticle distribution and wetting properties of the par-
ticulate being removed.  Consequently, contact power
correlation is used in this report only to demonstrate
that scrubber conditions were typical from the viewpoint
of historical performance, if not from design condi-
tions, and to provide a data point which identifies
an average scrubber performance of 99.3-percent effi-
ciency with a contact power of 6.6 HP/1000 SCFM for
filterable particulate matter according to the nine
particulate tests conducted.

-------
                 - 60 -
6.2.3  System Performance

       The following calculationa1 procedure is used in
       computing the estimates of  total system abate-
       ment efficiency.   From 16 to 24 coke pushes were
       included in any given particulate test.  The
       characteristics of each push varied depending
       upon the process  parameters.  The tables refer-
       enced and described in Sections 6.1.2 and 6.1.3
       are used to determine average process and engineer-
       ing data and average emissions data, respectively,
       for the  system abatement efficiency computations.

       To obtain an estimated range of hood capture effi-
       ciency,  it is necessary to  estimate average coke
       push greenness.  The engineer-observer's estimate
       of "average degree-of-greenness" is used as the
       criterion for determining the overall average
       greenness.  From  this measure of overall greenness
       and from the two  regression lines determined
       from fugitive emission measurements and visual
       estimates, two estimates of hood capture efficiency
       can be determined.  These hood capture efficien-
       cies establish a  range which accordingly determines
       a range  of estimated fugitive emissions and total
       system efficiencies in accordance with the mass
       balances.  With scrubber inlet particulate test
       results, fugitive emissions can be calculated as
       shown above.  The sum of fugitive emissions and
       scrubber inlet emissions  comprises an estimate
       of the total coke-push emissions during the par-
       ticulate test (see Equation (2) in Section 6.2.1).
       As a check on this sum, it  is noted that the coke-
       push emissions should be equal to the scrubber
       inlet emissions divided by  the hood efficiency
       of capture.  With measurements of stack emission
       rates from the particulate  test, the total effi-
       ciency is computed from the sum of the stack and
       fugitive emission rates divided by the coke-pushing
       emission rate.

       As a check on this total system efficiency com-
       putation, the multiplication of hood capture
       efficiency and scrubber efficiency should yield
       this total system efficiency.  This procedure is
       utilized for each of the nine particulate tests
       and for  any other sampling  tests where both
       simultaneous scrubber inlet and outlet emission
       rates were measured.

       For each type of  test, then, overall averages of
       hood capture efficiency, scrubber inlet emissions,
       fugitive emissions, stack emissions, and total

-------
                      - 61 -
            efficiencies are computed.  These overall averages
            are used in evaluating the performance of the total
            system in abating pushing emissions.  The results
            obtained are contingent upon the "model" assump-
            tions which include:

                 1.  The engineer-observer's estimate of
                     the degree-of-greenness is sufficient
                     to estimate  the hood capture efficiency
                     quantitatively by two independent methods;

                 2.  The regression lines obtained from the
                     fugitive emission measurements and visual
                     emission measurements establish a range
                     of estimated hood capture efficiencies
                     that contains the actual hood capture
                     efficiency;

                 3.  Scrubber inlet and outlet emission rates
                     were simultaneously measured; and

                 4.  The estimate of hood capture efficiency
                     as obtained  from particulate matter can
                     be extrapolated to determine overall
                     gaseous contaminant abatement efficien-
                     cies .
6.3  Evaluation of Hood Capture Efficiency

     Section 5 .2 .13 discussed the mathematical basis for the
     method (Method 1)  used to estimate hood capture effi-
     ciency from high-volume air sampling data.  This section
     discusses the data reduction and correlation procedures
     used to derive hood capture efficiency estimates from
     the data  base presented in the tables in Appendix I and
     Tables 6.1.4-1 through 6.1.4-5.

     As indicated previously, an engineer-observer was present
     to accumulate certain process parameters and hood cap-
     ture efficiency estimates during a large portion of the
     source testing phase of the program.  The estimates by
     the engineer-observers comprise  the largest data base,
     by far, from which visual hood capture efficiencies can
     be deduced.  In addition, however, it was necessary to
     obtain some measure of the fugitive emissions, in accord-
     ance with previously mentioned procedures, to provide a
     quantitative estimate of the hood capture efficiency.  To
     relate these emissions measurements to the degree of coke
     push greenness, photographs were taken at the beginning,
     middle, and end of the coke pushes.  In addition, a fugi-
     tive emissions observer rated the particular coke push.
     The engineer-observer provided a second, independent

-------
                 - 62 -
observation of the coke push greenness.  Supplementing
these judgments of coke push conditions, Ford Motor
Company personnel also routinely record whether each
coke push is "green" or "clean" as it emerges from the
oven.

The photographs were rated independently by two analysts,
each grading a coke push as "clean," "medium," or "green."
Thus, five independent estimates of the degree of coke
push greenness were obtained.  These "degrees-of-green-
ness" were rated as 1, 2, and 3, corresponding respect-
ively to "clean," "medium," and "green."

During the analysis of these five independent observa-
tions, considerable disagreement was noted among the ob-
servers.  In order to determine whether any one observer
was biased, each observer's readings were compared statis-
tically to the average of all other observers' readings
using the non-parametric "Sign test."'  '   The only ob-
server found to be biased was the Ford Motor Company
observer.  Although comparison of the Ford observer with
the engineer-observer was possible over a large data
base (that is, the entire several months of data avail-
able from Ford records and engineer-log records), there
was no reason to believe that the engineer-observer was
not also biased.  Therefore, the bias test was conducted
only using a data base where a minimum of three other
observers could be averaged in an attempt to determine
bias of any fourth observer.  As a result, all Ford
observer readings were excluded from subsequent analyses.

The next step in the determination of hood capture ef-
ficiencies was the averaging of the non-biased observers'
readings to determine an average greenness for each push
that was characterized by filtration measurements and
photography.  In ord'er to correlate the suspended dust
concentration measured above the hood with average green-
ness value, three operational codes were established
and defined as follows:

     Code 1 - Hood not in place and exhaust fan not
              running

     Code 2 - Hood in place and exhaust fan not running

     Code 3 - Hood in place and exhaust fan running.

For each of these individual codes, the logarithms of
concentrations obtained by filtration were correlated
with the average greenness computed from the four un-
biased observers' estimates.  The regression lines were
plotted on semi-log paper and 95-percent confidence
intervals were established.  Any data points outside  the
95-percent confidence  interval  and any  data  points associated

-------
                 - 63 -
with photographs deemed atypical (in the sense that the
plume appeared to pass the high-volume air sampler in such
a way as to not be sampled representatively) by two or
more independent analysts, were excluded from the data
base for that particular code.  New regression lines were
then computed using the revised data base.  Figures
6.3-1 to 6.3-3 display the semi-log regression estimates
of concentrations above the hood as a function of average
greenness for the three system configurations.  Those data
points deemed to be "outliers" according to the aforemen-
tioned criteria are so indicated in Tables 6.1.4-1 to
6.1.4-5 which list the pertinent parameters comprising
fugitive emissions data.

An initial analysis of variance indicated that the Code 1
and Code 2 fugitive emissions data might not be signifi-
cantly different from one another.  The Code 3 data,
however, was found to have a mean value that was signifi-
cantly different from the other two sets.  Thus, when the
regression lines were constructed the lines for Code 1
and Code 2 were tested for homogeneity.  Only the test
for homogeneity of variances was completed.  Since the
residual variances were found to differ significantly,
no further testing was required and the two lines were
considered  unique.  The analogous computation was not
carried out using the Code 3 line because of the results
of the analysis of variance and the differences apparent
upon visual inspection of the plots.

As indicated in Section 5.2.13,the basic difference be-
tween concentrations measured above the hood for Codes 1
and 2 rests in the contribution of the hood in dispersing
fugitive emissions and providing a more diffuse plume.
Therefore, based upon the above tests, the regression
lines obtained for Codes 1 and 2 were considered dis-
tinguishable, indicating that the fugitive emissions
measurement technique is sensitive enough to discern
the effect of the non-exhausted hood on fugitive plume
concentrations of filterable particulate.  Therefore,
the data of Codes 1 and 2 were not combined.

To provide a more meaningful estimate of hood capture
efficiency, hood capture efficiency was defined as (see
Equation (3) in Section 6.2.1):


     g   =  1 _ concentration for Code 3
      H         concentration for Code 2*
The concentration measured during a Code 2 push (for
a given greenness rating) is considered to reflect
the total loss of all push emissions since the exhaust

-------
                      - 64 -
     fan is off.  Using this definition, Figure 6.3-4 displays
     the variations in estimated hood capture efficiency cal-
     culated, from Codes 2 and 3 as a function of average coke
     push greenness.

     To obtain a second estimate of hood capture efficiency,
     regression analysis was used to relate the hood capture
     efficiency visually estimated by the engineer-observer
     to average greenness.  The average greenness used in this
     correlation was the same average greenness used in the
     previous correlation, and the estimated hood capture
     efficiency was that obtained directly from the hood cap-
     ture data sheets (Appendix G).  This estimate of capture
     efficiency was used in an attempt to fit a regression
     line which relates estimated efficiency as a direct
     function of average greenness.  The results of this re-
     gression analysis are also shown in Figure 6.3-4.

     These two techniques provide a range of estimated hood
     capture efficiencies established by two independent cal-
     culational methods using an identical data base.  They
     provide the best estimate available as a result of this
     study to define a range in estimated hood capture effi-
     ciency and estimated total emissions and estimated total
     system abatement efficiencies.  As indicated in Section
     6.2.1, the total system abatement efficiency, determined
     from hood capture eifficiency and scrubber removal effi-
     ciency, is limited in accuracy to the extent that the
     hood capture efficiency is incorrectly depicted by the
     two fugitive emissions estimates.  This qualification
     applies as well to all computed total and fugitive emis-
     sion factors.  Nevertheless, although the fugitive emis-
     sions estimates are based upon methods that are largely
     unproven (see Section 5.2.13), we believe the capture
     efficiencies predicted by the two regressions to be
     reasonable approximations of actual hood performance.


6.4  Emission and Process/Control System Correlations

     It is useful and informative to devise predictive or
     correlative models that relate pushing emissions to pro-
     cess or control system parameters.  If highly-significant
     correlations are obtained, the results can be used, within
     limitations, to extend findings in this study to other
     coke batteries that contemplate application of the same
     or similar control systems.

     Based upon experience and subjective considerations, it
     is informative to examine the following relationships:

          1.  Control device stack gas opacity as a function
              of greenness rating of the push;

-------
                 - 65 -
     2.  Filterable particulate pushing emission rate
         as a function of the average greenness rating;

     3.  Particulate size distribution of pushing emis-
         sions as a function of average greenness rating;

     4.  Hood capture efficiency as a function of average
         greenness rating;

     5.  Total system abatement efficiency as a function
         of average greenness rating;

     6.  Filterable particulate pushing emission rate
         as a function of the ratio of number of green
         pushes to number of clean pushes;

     7.  Greenness rating of the push as a function of
         net coking time;

     8.  Greenness rating of the push as a function of
         coking temperature;

     9.  Particulate size distribution of pushing emis-
         sions as a function of the organic material
         content of the particulate;

    10.  Hood capture efficiency as a function of ambient
         wind velocity;

    11.  Filterable particulate pushing emission rate
         (measured in a multi-push particulate test)
         as a function of average net coking time; and

    12.  Filterable particulate emission rate as a function
         of oven temperature.

Some of these correlations have been attempted in this
study.  However, several could not be attempted for vari-
ous reasons.  The "printer-clock" at the Ford "A" and
"B" Battery control room, which automatically records
the net coal charged and the charge time, was malfunction-
ing for several months.  Although the net coal charge
weights were recorded, the charge times were not.  Con-
sequently, the gross coking time (i.e., push-to-push time
for a given oven), rather than net coking time, is the
only parameter available for correlations based upon
this independent variable.

The particle sizing method used, the Andersen impactor,
provided too little particulate matter for determination
of organic materials content.  Consequently, the ninth
correlation listed could not be attempted.  The results
of all other correlations are included.  Section 7.0

-------
                 - 66 -
presents those results which relate to system performance,
including hood capture efficiencies and scrubber effi-
ciencies for specific materials.

It should be noted that correlations deemed "significant"
herein based upon their respective correlation coeffi-
cient do not indicate a cause-and-effect relationship; ,
they only imply that some relationship exists.  In actu-
ality, a third factor may be affecting both of the other
two factors and causing the apparent correlation.

Average scrubber outlet stack gas opacity correlates well
with average greenness rating when the nine multi-push
particulate tests are used.  The correlation coefficient
is 0.843, significant at the 99-percent level.

The filterable particulate pushing emission rates (i.e.,
fugitive emissions plus those captured by the hood) appar-
ently correlate well, with the average greenness ratings.
The two methods of calculating pushing emission rates
provide two separate estimates of the relationship.  For
Method 1 (see Section 6.2.1), the correlation coefficient
is 0.819, significant at the 99-percent level.  Using
Method 2 to estimate hood capture efficiency, the correla-
tion coefficient is 0.859, also significant at the 99-
percent level.

The inlet particle size data provide useful information
required to design and assess the control device used
to abate collected emissions.  As such, correlations
were attempted to relate an index of the inlet particle
distribution to the average greenness rating of pushes
that comprised each test.  It was found that, for 19
data sets, the percentage of inlet particles on a mass
basis less than one micron in diameter are not correlated
significantly with greenness.  However, the inlet per-
centages of particles, on a mass basis, less than five
microns and less than 10 microns were correlated signifi-
cantly with greenness (at the 95-percent confidence level)
as evidenced by their respective correlation coefficients
of 0.462 and 0.518.  Each correlation used the inlet par-
ticle size data sets listed in Table 7.3.1-15.

The correlation between hood capture efficiency and average
greenness rating has been presented previously in Section
6.3.  The results have been plotted in Figure 6.3-4.

Total system abatement efficiency apparently correlates
well with average greenness rating for filterable par-
ticulate matter.  Total system efficiency is the product
of hood capture efficiency and scrubber efficiency (see
Section 6.2.1 and Equation (8) in Section 6.2.1).  Using

-------
                 - 67 -
Method 1 to estimate hood capture efficiency, the correla-
tion coefficient is -0.998, significant at the 99-percent
level.  Using Method 2 to estimate hood capture efficiency,
the correlation   coefficient is -0.999, significant at
the 99-percent level.

An apparently significant correlation was found between
the filterable particulate emission rate (measured in a
multi-push particulate test) and the ratio of the number
of green-to-clean pushes constituting the test.  Again,
two results were obtained, depending upon which method
was used to estimate hood capture efficiency.  Using
Method 1, the correlation coefficient is 0.876, signifi-
cant at the 99-percent level.  Using Method 2, the correla-
tion coefficient is 0.912 which is significant at the
99-percent level.

An apparently significant correlation was found between
the greenness estimate for a given push (documented by
the engineer-observer) and the gross coking time.  The
data, consisting of 320 pairs of greenness ratings and
gross coking times, exhibit a correlation coefficient of
-0.157, significant at the 99-percent level.

An apparently significant correlation was also found be-
tween greenness rating and oven temperature (recorded
once each shift; see Section 5.1.5).  The correlation
coefficient for the 164 data pairs obtained during the
nine particulate tests is -0.374, significant at the
99-percent level.

An apparently significant correlation was also found be-
tween greenness rating and the product of gross coking
time and oven temperature.  The correlation coefficient
for the 207 sets of data is 0.331, significant at the
99-percent level.

No apparent linear correlation was found between hood
capture efficiency and wind velocity as measured and
documented by the on-site observer.  The measure of hood
capture efficiency used in this attempted correlation
was that recorded by the engineer-observer from his vis-
ual estimate (i.e., Method 2).  These data are presented
in Appendix G.

No apparent linear correlation exists between the filter-
able particulate push emissions measured in the nine
particulate tests (and expressed as pounds per ton of
coal charged) and average gross coking time.  The correla-
tion coefficients of -0.502 and -0.505 for Methods 1 and
2, respectively, (see Section 6.3), are not significant.
This is likely attributable to the use of gross coking

-------
                 - 68 -
time, rather than net coking time, as the independent
variable.  Furthermore, because the particulate tests
were necessarily multi-push tests, the identity of the
emissions from an individual push is "diluted" by the
others.  This is further complicated by the narrow
range in gross coking time, 940 to 1053 minutes.

No apparent linear correlation exists between the fil-
terable particulate emission factors and the average
oven temperature for the ovens pushed during each of the
nine multi-push particulate tests.  The correlation co-
efficients of 0.320 and 0.349 for Methods 1 and 2, re-
spectively, are nonsignificant.  Again, because the
particulate tests were necessarily multi-push tests, the
identity of the emissions from an individual push is
"diluted" by the others.  Furthermore, oven temperatures
are recorded only once per shift.

We emphasize that the correlations presented above are
based upon limited data.  While the results obtained
are significant statistically, the extent to which pro-
cess emissions are predictable by considering only single
independent variable correlations is open to question.
The basis used in prescribing the variable to be correlated
and the mathematical form of the correlation is largely
intuitive and is based upon our experience in other coke
oven emission studies.  It is possible that multiple re-
gression type correlations may reveal statistically sig-
nificant relationships if the independent variables are
judiciously chosen from the available data base.  Attempts
at such correlations are recommended and the requisite
data are available in the aforementioned appendices of this
report.

-------
                             - 69 -


                    7.0  SUMMARY OF FINDINGS

This section presents an evaluation by Clayton Environmental Con-
sultants, Inc. of the Ford/Koppers coke-pushing emission control
system based upon:

     1.  Overall control system performance and reliability;

     2.  Estimated capture efficiency of the exhaust hood device
         during coke pushing; and

     3.  Scrubber abatement efficiencies for both gaseous and
         particulate contaminants.

The process data and emission measurements for each test are summa-
rized in detail in Section 6.0.  Tables 7.0-1 through 7.0-14 relate
these two types of measurements and present:  1) contaminant emis-
sions measured at the inlet and outlet of the scrubber, 2) scrubber
control efficiency, 3) estimated hood capture efficiency (as deter-
mined by each of two methods), 4) fugitive emissions as estimated
from hood capture efficiency, 5) total estimated pushing emissions
(computed from the sum of inlet emission to the control system and
the hood fugitive emissions), and 6) estimated overall control sys-
tem efficiency (based upon the two methods of estimation of hood
capture efficiency).

Although the accuracy of the hood capture efficiency cannot be estab-
lished using "proven" techniques, the precision .of the statistically
estimated capture efficiencies (by regression techniques) is good.
For the set of nine multi-push particulate tests, the mean estimate
of hood capture efficiency by Method 1 is 41 percent (+ 5 percent)*;
the mean estimate by Method 2 (visual judgment) is 66 percent (+ 9
percent)*.                                                     ""

Despite the precision of each method, the range of capture efficien-
cies defined by Method 1 and Method 2 suggests that the emission
factors presented in Tables 7.0-1 through 7.0-14 need to be quali-
fied as estimates that have not been verified by independent meas-
urement .

Some of the contaminants listed' in Table 3.2-2 are excluded from
scrubber efficiency and mass emission rate analysis because their
measured values were found to be consistently below the limits of
detectability.

Accordingly, the contaminants presented in Tables 7.0-1 through
7.0-14 include:  filterable particulate and condensibles, the chloro-
form-ether soluble fraction of filterable particulate and conden-
sibles, sulfur dioxide, sulfur trioxide, sulfur oxides (as sulfates) ,
carbon monoxide, nitrogen oxides (as NC>2) , benzene, benzene and
homologues, total light hydrocarbons (as methane), methane and homo-
logues, and ethylene and homologues.
* The notations (+ 5 percent) and (+ 9 percent) are estimates of
  the precision of the mean values "based upon a 95-percent level
  of confidence.

-------
                             -  70  -
Conversion factors to enable computation of emission factors in
terms of dry coal charged are presented in Section 1.2 of this
report.


7.1  Overall System Performance

     Generally, the aspect of system performance which most directly
     determines and affects the system evaluation is the ability
     of the hood and scrubber to capture and remove emissions.  In
     addition to efficiency, certain operating characteristics must
     be considered in this evaluation to determine the frequency
     with which the abatement system performed as intended.

     Ideally, records of the number of times the emission control
     system came to the full operation condition would indicate
     the percentage of the time that the emission control system
     functioned properly.  Unfortunately,  charts in the control  room
     which record fan amperage (and therefore the degree of system
     operability) were not saved and, therefore, could not be re-
     searched to determine long-term operability by this method.
     However, during the months of June and July, 1975, the engineer-
     observer and the test crew present during the day shift re-
     corded the number of times the system functioned at full operat-
     ing potential.  Various malfunctions, which included the failure
     of the  Rayco  cylinder to open completely the fume main doors
     and thereby actuate the inlet fan louvers, consistently con-
     tributed to the temporary failures of the system.  Table 6.1.1-3,
     as previously presented, lists those  failures considered to be
     chronic.

     Any assessment of the emission control system which considers
     total abatement efficiency must also  include the fraction of
     the time for which the system functioned according to design.
     During those days when Clayton observers were on-site near  the
     end of June, 1975, approximately half of the pushes during  the
     day shifts resulted in totally uncontrolled emissions.  Sys-
     tem observation during other parts of June and during the
     month of July, 1975, indicated that the system performed nearly
     80 percent of the time to abate pushing emissions consistent
     with control efficiency estimates presented in Tables 7.0-1
     through 7.0-14.

     The "downtime" observed during this period was unusual based
     on the system observation conducted intermittently during the
     previous year.  The chronic failure of the Rayco cylinder
     caused the main fan louvers to not open, leaving the control
     system in idle mode during many pushes.  Once the cause was
     found and corrected, the system returned to its normal operabil-
     ity.  Table 6.1.1-3 shows clearly that this unusual period  of
     chronic failure  does not necessarily  represent a gradual degrad-
     ation of operability or availability.

-------
                             - 71 -
     During other months of 1974 and 1975, visual observations were
     made by the engineer-observer during the day shift to charac-
     terize the hood capture efficiency and stack opacity and summa-
     rize, in general, the system operability.   No conclusions can
     be made about the percentage of pushes controlled to any degree
     during the second and third shifts because' the Ford Motor Com-
     pany maintenance records indicate that repair of malfunctions
     was often scheduled for daytime maintenance crews.  However,
     frequent difficulty in diagnosing basic causes for system com-?
     ponent failure caused the system to fail consistently during
     some periods.

     Any attempt to cite a representative percentage of time for
     which the emission control system was down is necessarily bi-
     ased by the nature of the malfunction occurring while the
     engineer-observer was on-site.  However, such observations in-
     dicate that the system performed approximately 80 percent of
     the time in abating pushing emissions.  This conclusion is
     based upon system observations during the  months of August and
     September of 1974, intermittent observations during January,
     February, and April of 1975, and continuous observation during
     June, July, and portions of August and September of 1975.  Thus,
     a total, overall system performance evaluation should include
     multiplication of total abatement efficiencies by 0.8 to account
     for the 20-percent downtime estimated by Clayton observation.


7.2  Hood Capture Efficiency

     As previously indicated, Figure 6.3-4 displays the hood capture
     efficiency estimated by two different methods.  This figure
     indicates that for a push with an average  greenness of 1.0
     (clean push), the estimated hood capture efficiency ranges be-
     tween 54 and 92 percent.  Similarly, a push exhibiting an aver-
     age greenness of 2.0 results in a hood capture efficiency esti-
     mated to range between 38 and 61 percent.   A push which exhibits
     an average greenness of 3.0 (green push) leads to a hood cap-
     ture efficiency estimated to range between 22 and 30 percent.
     Because several coke pushes were included  in some of the par-
     ticulate and gaseous contaminant tests, it was necessary to
     resort to two-significant-figure accuracy  to compute average
     coke-pushing greenness.  Interpolation was used in Figure 6.3-4
     to provide a range of hood capture efficiencies.

     The results reported in this section are for those contaminants
     which exhibited concentrations which were  greater than the
     limit of detectability.  Of the other contaminants listed in
     Table 3.2-2, the following were not included in Tables 7.0-1
     through 7.0-14 because values reported were all below the limit
     of detectability:  acetylene, acridine, acrolein, ammonia, anthra-
     cene, benzo(a+e)pyrene, carbazole, carbon  disulfide, chlorine,
     chrysene, cyanide, fluoranthene, fluorene, 9-fluorenone, hydro-
     gen sulfide, naphthalene, phenanthrene, pyrene, pyridine, total

-------
                        - 72 -
phenolics, and total reduced sulfur.  The process conditions
existing during tests for these compounds and the calculated
mass emission rates are expressed in the data listed in Appen-
dix I and Tables 6.1.3-1 through 6.1.3-92, respectively.
The accuracy of these emission factors is contingent upon the
accuracy of the estimated fugitive emissions.

As indicated, emission factors are presented, in this section,
in English dimensional units as this system is most widely used
in the coking industry.
7.2.1  Particulate Matter

       During the nine particulate tests, average greenness
       ratings ranged from 1.4 to 2.4, with an arithmetic
       mean of 1.8.  Hood capture efficiencies, as estimated by
       fugitive emission concentration measurements  ( Method
       1 ), during these nine tests ranged from 32 to 48 and
       averaged 41 percent.  For these same nine particulate
       tests, the hood capture efficiency estimated  from the
       engineer-observer's visual observations ( Method 2 )
       ranged from 49 to 80 percent with an average  of 66 per-
       cent.  Therefore, as shown in Table 7.0-1, it can be
       concluded that the hood captures between 41 and 66 per-
       cent of all filterable particulate coke-pushing emis-
       sions, assuming the nine particulate tests are charac-
       teristic of coke-pushing emissions normally produced
       at Batteries A and AX.  For these ranges of greenness
       ratings and hood capture efficiencies, fugitive fil-
       terable particulate emissions, expressed as "pounds per
       ton of coal charged," ranged from 1.07 to 2.55 and
       averaged 1.70 as estimated by Method l;and from 0.246
       to 1.25 with an average of 0.629 as estimated by Method
       2.  Similarly, for the same set of nine particulate
       tests, fugitive condensible emissions ranged  from 0.0101
       to 0.0982 and averaged 0.0508 as estimated by Method 1;
       and from 0.0023 to 0.0481 with an average of  0.0199
       as estimated by Method 2 (see Table 7.0-2).

       In an attempt to determine the organic fraction of fil-
       terable and total particulate emissions, analytical
       procedures were employed which use chloroform-ether ex-
       traction techniques (see Section 5.2.1).  The materials
       soluble in chloroform-ether are thus defined  in this
       study as organic constituents of the sampled  particulate.
       Tables 7.0-3 and 7.0-4 display, respectively, the filter-
       able and condensible chloroform-ether .soluble particulate
       emissions and control efficiencies.    Because the same
       samples were analyzed for these "organic" constituents
       as those analyzed and reported in Tables 7.0-1 and 7.0-2,
       the greenness ratings and hood capture efficiencies are
       identical.  Fugitive emissions of chloroform-ether solu-
       ble, filterable particulate, based on Method  1, ranged

-------
                        - 73 -
       from 0.0023 to 0.0536 and averaged 0.0211 pound  per ton
       of coal charged (see Table 7.0-3).  These same emissions
       estimated by Method  2 ranged  from 0.0005  to 0.0244 and
       averaged 0.0084 pound per ton of coal  charged.  Fugitive
       chloroform-ether soluble, conderisible  emissions  ranged
       from <0.0005 to 0.0100 and averaged.0.0046 to 0.0047
       pound per ton of coal charged as estimated by Method 1;
       and from <0.0002 to,0.0049 with an average of 0.0017
       pound per ton of coal charged as estimated by Method 2
       (see Table 7.0-4) .


7.2.2  Sulfur Oxides

       Tests designed to measure sulfur dioxide  and sulfur tri-
       oxide indicated hood capture  efficiencies computed by
       Method 1 ranging from 22 to 43 with an average of 38 per-
       cent.  For the coke  pushes tested, averaged greenness
       ratings ranged from  1.7 to 3.0 with a  mean of 2.0.  Cor-
       respondingly, the average hood capture efficiency esti-
       mated by Method 2 ranged from 30 to 71 with a mean of
       61 percent.  The emission control system  thus captures
       bewteen 38 and 61 percent of  all sulfur dioxide  (see
       Table 7.0-5) and sulfur trioxide emissions (see  Table
       7.0-6), assuming that the coke pushing emissions during
       this series of tests are typical of those emitted over
       long periods of time.  The average hood capture  efficien-
       cies for sulfur dioxide and sulfur trioxide are  assumed
       to be identical for  each test because  these contaminants
       are determined by analysis of samples  from the same im-
       pinger contents. However, fugitive emissions of sulfur
       dioxide are substantially different from  sulfur  trioxide
       because they are based upon the total  mass of each species
       emitted.

       The fugitive sulfur  dioxide emissions  expressed  as pounds
       of sulfur dioxide per ton of  coal charged ranged from
       0.0095 to 0.0418 and averaged 0.0166 as determined by
       Method 1.  Similarly, sulfur  dioxide emissions ranged
       from 0.0029 to 0.0275 and averaged 0.0077 as determined
       by Method 2.

       Fugitive sulfur trioxide emissions expressed as  pounds
       of sulfur trioxide per ton of coal charged ranged from
       less than 0.00057 to 0.014 and averaged 0.0019 to 0.0030
       as determined by Method 1.  Similarly, sulfur trioxide
       emissions based on Method 2 ranged from less than 0.00018
       to 0.0056 and averaged 0.0007 to 0.0013.

       Total sulfur oxides  measured  as "sulfate" are tabulated
       in Table 7.0-7.  For the pushes tabulated, the hood cap-
       ture efficiency ranged from 22 to 54 percent with an
       average of 36 percent using Method 1.   Method 2  yielded

-------
                        - 74 -

       hood capture efficiencies ranging from 30 to 92 with
       an average of 58 percent.  Therefore, the hood capture
       efficiency (see Table 7.0-7) ranged from 36 to 58 percent
       assuming that the pushes sampled produced typical sul-
       fur oxide emissions.  Fugitive emissions of sulfur oxides
       expressed as pounds of sulfate per ton of coal charged
       ranged from 0.0354 to 0.465 and averaged 0.176 using
       Method 1.  Similarly, fugitive sulfur oxides emissions
       ranged from 0.0036 to 0.220 with an average of 0.089 us-
       ing Method 2.  These six tests included pushes which ex-
       hibited average greenness ratings ranging from 1.0 to
       3.0 with a mean of 2.1.
7.2.3  Carbon Monoxide

       For the six carbon monoxide tests included in the sum-
       mary (see Table 7.0-8), the hood capture efficiency ranged
       from 22 to 54 percent with an average of 44 percent using
       Method 1.  Using Method 2, the efficiency ranged from
       30 to 92 and averaged 73 percent.  The average greenness
       ratings during these six grab-sample tests ranged from 1
       to 3 with a mean of 1.6.  Fugitive emissions of carbon
       monoxide expressed as pounds of carbon monoxide per ton
       of coal charged for Method 1 ranged from less than 0.0027
       to 0.074 with an average of 0.032 to 0.033.  Uncaptured
       emissions of carbon monoxide based upon Method 2, ranged
       from less than 0.00028 to 0.049 with an average of 0.015.
7.2.4  Oxides of Nitrogen

       Hood capture efficiencies for the 13 nitrogen oxides te sts
       ranged from 22 to 54 with an average of 44 percent using
       Method 1.  Similarly, the capture efficiencies  for nitro-
       gen oxides using Method 2 ranged from 30 to 92  and averaged
       73 percent.  These estimates are based on the average green-
       ness ratings for these same 13 tests ranging from 1 to
       3 with a mean of 1.6.  The hood capture efficiencies thus
       ranged from 44 to 73 percent assuming that these 13 pushes
       are typical of the nitrogen oxides emissions.  Fugitive
       emissions of nitrogen oxides expressed as pounds of nitro-
       gen oxides per ton of coal charged range from less than
       0.0014 to 0.0603 with an average of 0.011 using Method 1.
       Uncaptured emissions of nitrogen oxides using Method 2
       ranged from less than 0.00015 to 0.0397 and averaged 0.0057
       pounds per ton of coal charged (see Table 7.0-9).


7.2.5  Benzene and Homologues

       Based upon the six test results for benzene emissions
       displayed in Table 7.0-10, the hood capture efficiency
       using Method 1 ranged from 22 to 43 and averaged 33 per-
       cent.  The average greenness ratings for these  six tests

-------
                        - 75 -
       ranged from 1.7 to 3 and averaged 2.3.  Similarly, hood
       capture efficiencies using Method 2 ranged from 30 to
       71 with an average of 51 percent.  Fugitive emissions
       of benzene using Method 1 expressed as pounds of ben-
       zene per ton of coal charged ranged from less than
       0.0001 to 0.007 with an average of 0.002.  Fugitive
       emissions based on Method 2 ranged from less than 0.00003
       to 0.005 with an average of 0.001 pound per ton of coal
       charged.  Four of the six tests reported less-than
       values.  These data indicate that these uncaptured emis-
       sion estimates must be considered approximate.

       Laboratory tests conducted for benzene and homologues
       emissions are based upon analytical measurements made
       from the same charcoal tubes from which benzene ana-
       lytical results were obtained.  Average greenness ratings
       and hood capture efficiencies   are the same as those re-
       ported for benzene emissions.   Fugitive emissions based
       upon Method 1 expressed as pounds of benzene and homo-
       logues per ton of coal charged' range from 0.0010 to
       0.132 and average 0.025.  The  fugitive emissions based
       on Method 2 range from 0.00032 to 0.0868 and average
       0.016.  None of the values for benzene and homologues
       reported for these six tests were less-than values (see
       Table 7.0-11).


7.2.6  Total Light Hydrocarbons

       Average greenness ratings for  the grab-sample tests for
       total light hydrocarbons (as methane)  ranged from 1 to
       3 with a mean of 1.5.  Hood capture efficiency for total
       light hydrocarbons based upon  Method 1 ranged from 22 to
       54 with an average of 46 percent.  Hood capture effi-
       ciencies based upon Method 2 ranged from 30 to 92 and
       averaged 76 percent,  Fugitive emissions for total light
       hydrocarbons expressed as pounds of total light hydro-
       carbons per ton of coal charged based  upon Method 1
       ranged from 0.0005 to 0.0825 with an average of 0.017.
       Fugitive emissions based on Method 2 ranged from 0.00005
       to 0.00842 and averaged 0.0026.  None  of the values
       reported for the six tests for total light hydrocarbons
       are less-than values (see Table 7.0-12).
7.2.7  Methane and Homologues

       Methane and homologues emissions (see Table 7.0-13)
       were obtained from the same samples as were the results
       for total light hydrocarbons.   The average greenness
       ratings and hood capture efficiencies are the same as
       those delineated above for total light hydrocarbons.
       Fugitive emissions based on Method 1 expressed as pounds

-------
                             - 76 -
            of methane and homologues per ton of coal charged ranged
            from 0.0003 to 0.003 and averaged 0.001.  The fugitive
            emissions ranged from 0.00003 to 0.0009 and averaged
            0.0003 based on Method 2.


     7.2.8  Ethylene and Homologues

            The samples analyzed for ethylene and homologues were
            those used for the methane and homologues and the total
            light hydrocarbon tests.  Average greenness ratings and
            hood capture efficiencies are the same as those deline-
            ated above for total light hydrocarbons.  Fugitive emis-
            sions based on Method 1 expressed as pounds of ethylene
            and homologues per ton of coal charged ranged from less
            than 0.0002 to 0.0516 and averaged 0.010.  Fugitive
            emissions based on Method 2 ranged from less than 0.00002
            to 0.00527 and averaged 0.001.  The values of ethylene
            and homologues reported for two of the six tests were
            less-than values (see Table 7.0-14).

     In summary, hood capture efficiencies as determined by the two
     methods ranged on the average between 33 and 46 percent as
     measured by Method 1, and between 51 and 76 percent as measured
     by Method 2 for the series of tests conducted for each of the
     contaminants mentioned above.  General indications are that for
     the types of pushes sampled during the June and July test peri-
     ods, the hood capture efficiencies averaged approximately 50
     percent or slightly greater.  These efficiencies are contingent
     upon the types of pushes encountered during the test period
     (see Section 8.3).

     Clayton Environmental Consultants believes that the reported
     ranges of emission factors (determined by Methods 1 and 2 for
    , estimating fugitive emissions) include a useful estimate of
     pushing emission factors for each species tested.  Nevertheless,
     we emphasize that the emission factors include a non-standard
     technique for estimating fugitive emissions and, as such, must
     be qualified as being limited in accuracy by the accuracy of
     the methods.


7.3  Scrubber and Total System Efficiencies

     Tables 7.0-1 through 7.0-14, as indicated previously, tabulate
     mass emission, rates into the collection system, fugitive emis-
     sion rates, and stack emission rates.  From these, total push-
     ing emissions are estimated.  Average hood capture efficiencies
     and scrubber efficiencies are calculated; the total system
     abatement efficiency is the product of these two efficiencies.
     The previous section provided estimates of hood capture effi-
     ciencies for each of the species tested for which some concen-
     trations were greater than the limit of detectability.  This

-------
                        - 77 -
section lists measured scrubber efficiencies and estimated
total system efficiencies for each of these species.

As indicated, emission factors presented in this section utilize
the English system of dimensional units since this system is
most widely used in the coking industry.
7.3.1  Particulate Matter

       For the nine particulate tests conducted, average green-
       ness ratings ranged from 1.4 to 2.4 with a mean of 1.8.
       For these tests, the scrubber efficiency ranged from
       99.1 to 99.6 and averaged 99.3 percent for filterable
       particulate; for condensibles, scrubber efficiency ranged
       from zero to 90.5 and averaged 47.7 percent.  The result-
       ant estimated overall system control efficiency for fil-
       terable particulate ranged from 32 to 48 and averaged 41
       percent based on Method 1; total system abatement effi-
       ciency for condensibles based upon Method 1 ranged from
       zero to 39 and averaged 20 percent.  The resultant esti-
       mated overall system control efficiency for filterable
       particulate emissions based upon Method 2 ranged from
       49 to 80 and averaged 66 percent; total system efficiency
       for condensibles based on Method 2 are estimated to range
       from zero to 64 and average 33 percent.  The estimated
       overall system abatement efficiency for filterable par-
       ticulate emissions averaged between 41 and 66 percent
       based upon the two methods used to estimate hood capture
       efficiency.

       If filterable "organic" particulate matter is defined as
       the chloroform-ether soluble fraction of filterable par-
       ticulate, then scrubber efficiency for chloroform-ether
       soluble filterable particulate ranges from 52.6 to greater
       than 98.4 and averages 87.2 percent, while estimated total
       system abatement efficiency for this material ranges
       from 35 to 58 percent.

       Particle sizing test results are summarized in Tables
       7.3.1-1 through 7.3.1-14.  The results are plotted on
       log-normal probability paper in Figures 7.3.1-1 and
       7.3.1-2 for the inlet and outlet, respectively.  For each
       set of lines, a line is constructed which averages
       the seven tests.  The average distribution is obtained
       by averaging the weight percentages obtained from each
       impactor stage for each set of inlet and outlet tests.
       The differences among the individual size distributions
       displayed in Figures 7.3.1-1 and 7.3.1-2 are likely
       attributable to variations in both process conditions
       and sampling technique.  The average line shows that the
       particle size distribution measured at the inlet displays
       a mean particle size diameter greater than 10 microns.

-------
                        - 78 -
       At the outlet, the mean particle size diameter is approx-
       imately 1.7 microns; the geometric standard deviation
       of the outlet distribution is 4.7.  All particle size
       tests were conducted with the Andersen impactor.

       The concentration of filterable particulate matter at
       the scrubber inlet and outlet has also been computed
       from the Andersen particle size samples.  The results
       are displayed in Table 7.3.1-15.  The ranges in calculated
       concentrations result from the use of "less than" values
       as equal to zero or equal to the less than value.  At
       the scrubber inlet, filterable particulate data (Table
       6.1.3-65  indicate concentrations ranging from 1.46 to
       1.91 grains per dry standard cubic foot (gr/DSCF).  Table
       7.3.1-15 indicates that inlet concentrations obtained
       from Andersen sampler data range from 0.551 to 1.34
       gr/DSCF.  At the scrubber outlet, filterable particulate
       concentrations ranged from 0.006 to 0.015 gr/DSCF (Table
       6.1.3-66 .  The Andersen sampler data indicate outlet
       concentrations ranging from less than 0.002 to 0.042
       gr/DSCF as shown in Table 7.3.1-15.  These results com-
       pare well considering that particulate emission tests
       and particle sizing tests were not conducted for the
       same set of pushes.  The inlet Andersen data in Table
       7.3.1-15 are computed from samples collected over a
       short time due to the high loading of particulate at
       that location.  The outlet data are based upon samples
       collected over much longer periods.  In most cases three
       inlet samples are taken for each corresponding outlet
       sample to provide a semblance of simultaneous testing.
       Test times are displayed in the table.

       Tables 7.3.1-16 and 7.3.1-17 list the components of the
       particulate matter collected in the nine inlet and out-
       let particulate tests, respectively.  The percentages of
       sulfate, iron, aluminum, titanium, and chloride are given.


7.3.2  Sulfur Oxides

       Scrubber control efficiencies for sulfur dioxide ranged
       from nil to 63 and averaged 42 percent.  Total system
       abatement efficiency based upon both estimates of hood
       capture efficiency ranged from nil to 45 percent.  The
       average- total system abatement efficiency for sulfur
       dioxide push emissions is estimated to range from 16 to
       26 percent.

       The average greenness of the nine sulfur trioxide tests
       ranged from 1.7 to 3.0 and averaged 2.0.  Sulfur trioxide
       scrubber efficiencies ranged from nil to greater than 91
       percent and averaged from 39 to 67 percent (depending
       upon the method of computation of the average).  The

-------
                        - 79  .
       resultant total system abatement efficiency for sulfur
       trioxide based upon the Method 1 estimation of average
       hood capture efficiency ranged from 9 to 42 percent and
       averaged 15 to 27 percent.  The total system abatement
       efficiency based upon Method 2 is estimated to range
       from 19 to 71 and average 25 to 44-percent.  Only, three
       of the nine tests permitted meaningful computation of
       scrubber efficiencies.  Therefore, the broad range of
       total system abatement efficiencies  must be considered
       approximate.  These conclusions are  largely dependent
       upon the low concentrations of sulfur trioxide measured
       at the inlet location.  For the nine tests conducted,
       six tests consistently exhibited less-than values  (that
       is, below the limit of detectability) for sulfur tri-
       oxide .

       For the six sulfur oxides tests conducted, the average
       greenness ratings ranged from 1.0 to 3.0 with an average
       of 2.1.  Scrubber control efficiencies for sulfur  oxides
       ranged from 41.6 to 80.1 and averaged 61.5.  The scrubber
       control efficiencies and average hood capture efficiencies
       yield estimates of total system abatement efficiencies
       for sulfur oxide emissions based on  Method 1 which ranged
       from 9 to 39 and averaged 23 percent.  Total system
       abatement efficiencies based on Method 2 are estimated
       to range from 13 to 66 and average 37 percent.  Estimated
       total system abatement efficiencies  for sulfur oxides
       thus ranged from 23 to 37 percent depending upon the
       method of computation of average hood capture efficiency.
7.3.3  Carbon Monoxide

       For the six carbon monoxide tests  conducted,  the average
       greenness ratings ranged from 1 to 3 and  averaged 1.6.
       Scrubber control efficiencies for  these  tests ranged
       from greater than 35 to greater than 84  and averaged
       greater than 54 percent.  Total system abatement effi-
       ciencies for carbon monoxide based  on Method  1 are
       estimated to range  from 13 to 54  and to  average from
       18 to 38 percent.  Total system abatement efficiencies
       based on Method 2 are estimated to range  from 21 to 92
       and to average  from 28 to 61 percent.  Total  estimated
       system abatement efficiencies for  carbon  monoxide ranged,
       therefore, from 18 .to 61 percent.   Only  three of the six
       tests conducted exhibited inlet emission  rates which
       were not less-than values.  No outlet emission rates
       for any of the  tests were greater  than the limit of de-
       tectability.  Therefore, these data must  be  considered
       approximate indications of total system  abatement of
       carbon monoxide emissions.

-------
                        - 80 -
7.3.4  Oxides of Nitrogen

       All 13 measurements for oxides of nitrogen made at the
       inlet and outlet of the scrubber indicated that outlet
       concentrations are equal to or greater than inlet con-
       centrations.  Therefore, control efficiencies were desig-
       nated as identically equal to zero.  As a result, total
       system abatement efficiencies for nitrogen oxides are
       zero by definition.


7.3.5  Benzene and Homologues

       For the six benzene emission tests conducted, the green-
       ness ratings ranged from 1.7 to 3 and averaged 2.3.  The
       scrubber abatement efficiencies for benzene emissions
       ranged from greater than 57 to greater than 85 and aver-
       aged greater than 71 percent.  Total estimated system
       abatement efficiency ranged from 19 to 33 percent and
       averaged from 21 to 28 percent using Method 1 for esti-
       mation of hood capture efficiency.  Total estimated sys-
       tem abatement efficiency based on Method 2 ranged from
       23 to 33 and averaged from 24 to 31 percent.   Total
       estimated system abatement efficiency for benzene ranges,
       therefore, from 21 to 31 percent.  Four of the six tests
       exhibited concentrations which were less than the limit
       of detectability.  Therefore, these results must be
       considered approximate abatement efficiencies for benzene
       emissions.

       The samples obtained for the six tests for benzene emis-
       sions were also analyzed for benzene and homologues.
       Scrubber control efficiencies for benzene and homologue
       emissions ranged from nil to 94 and averaged  34 percent.
       Total estimated system abatement efficiency based on
       Method 1 ranged from nil to 31 percent and averaged 11
       percent.  Total estimated system abatement efficiency
       based on Method 2 ranged from nil to 48 and averaged  16
       percent.  The total estimated system abatement efficiency
       of benzene and homologues, therefore, ranges  from 11  to
       16 percent based on these six tests.


7.3.6  Total Light Hydrocarbons

       For the- six tests conducted for total light hydrocarbons
       (expressed as methane), the greenness ratings ranged  from
       one to three and averaged 1.5.  Scrubber control effi-
       ciencies for total light hydrocarbons ranged  from nil to
       98 and averaged 23 percent.  Four of the six  tests indi-
       cated control efficiencies of zero.  Total estimated
       system abatement efficiencies for total light hydrocarbon
       emissions based on Method 1 ranged from nil to 53 and

-------
                         -  81  -
       averaged 12 percent.   Total estimated system abatement
       efficiencies ranged from nil to 90 and averaged  21 percent
       based on Method 2.   The  total estimated system abatement
       efficiency of total light hydrocarbons, therefore, ranges
       from 12 to 21 percent.


7.3.7  Methane and Homologues

       Samples analyzed for  methane and homologues  were the  same
       as those used to measure total light hydrocarbons.  For
       these six tests, scrubber control efficiencies for methane
       and homologues ranged from nil to 67 and averaged 17
       percent.  Total estimated system abatement efficiencies
       based on Method 1 ranged from nil to 33 and  averaged  8
       percent.  Total estimated system abatement efficiency
       based on Method 2 ranged from nil to 57 and  averaged  14
       percent.  The total estimated system abatement efficiency
       for methane and homologues, therefore, ranges from 8  to
       14 percent.


7.3.8  Ethylene and Homologues

       The test results for  ethylene and homologues were obtained
       from the same samples analyzed for methane and homologues.
       The scrubber control  efficiency for ethylene and homologues
       ranged from nil to  99.7  and averaged 39 percent.  Total
       estimated system abatement efficiency based  on Method 1
       ranged from nil to  54 and averaged 21 percent.  Total
       estimated system abatement efficiency based  on Method 2
       ranged from nil to  92 and averaged 36 percent.  The esti-
       mated total system  control efficiency for ethylene and
       homologues, therefore, ranges from 21 to 36  percent.


7.3.9  Summary of All Contaminants

       Table 7.3.9-1 summarizes the total estimated atmospheric
       emissions, including  hood fugitive emissions and scrubber
       stack emissions, for  those substances presented  in Tables
       7.0-1 through 7.0-14.  The table shows that  82 percent of
       all uncontrolled and  measurable emissions are particulate
       matter.  The remaining 18 percent consists of sulfur
       oxides, carbon monoxide, oxides of nitrogen, benzene  and
       homologues, total light  hydrocarbons, methane and homo-
       logues, and ethylene  and homologues.

       The other contaminants listed in Table 3.2-2 which were
       sampled at the scrubber  inlet and/or outlet  are  addressed
       in Table 7.3.9-2.  This  table lists the emission factors

-------
                 . 82 -
measured and expressed in terms of the limits of detect-
ability.  Where process data could not be obtained to
permit calculation of emission factors, the results are
expressed in concentration terms (parts per million).

-------
                             - 83 -


           8.0  DISCUSSION OF RESULTS AND CONCLUSIONS
This section presents a discussion of the investigations based
upon the data and results presented previously.  A summary of the
limitations of the study is also included.


8.1  Historical Performance Assessment

     Sections 5.1.1 and 5.1.2 identify the documents and process
     measurements that were utilized to evaluate the system with
     respect to its operability and maintainability.  One prime
     indicator is the required maintenance as documented in main-
     tenance logbook entries.

     While many system components contributed to control system
     downtime, only five major categories appeared consistently in
     the logbook.  Throughout the lifetime of the control system,
     the failure of the induced-draft fan and/or the main louvers
     accounted for prolonged periods of downtime.  The main fan
     failures, according to logbook entries, are traced to prob-
     lems with fan bearings.

     The failure of the induced-draft fan bearings contributed a
     large percentage of system downtime during which pushing
     emissions were emitted directly to the atmosphere.  The con-
     trol system, despite extensive maintenance, was not in opera-
     tion due to failures of the main fan or louvers for about 70
     days during the period 1972 to 1975.

     In researching the logbooks, it was hoped that any failures
     associated with seasonal variations would be obvious.  For
     example, one such indicator would be failure of the scrubber
     pumps due to freeze-up.  However, no discernible and consis-
     tent failure of the recirculation system pumps was observed.
     Approximately 15 failures of the pump were reported between
     1972 and 1975.

     Failure of the door machine due to its modifications to ac-
     commodate the control system was not a major contributor to
     system downtime.  Logbook entries indicate that the modified
     door machine experienced major failure 15 times throughout
     the system operation.  These failures did not appear to be
     seasonal.  Since the failures of the door machine cited in
     this report are malfunctions caused by modifications of the
     standard door machine and exclude downtime attributable to
     standard door machine performance, it appears that modifica-
     tion of existing door machines to house controls and sub-
     systems does not materially affect the standard door machine's
     performance since only about 15 failures are documented to
     have occurred.

-------
                         - 84 -
Malfunctions associated with either the door or the Rayco
cylinder, which actuates the control system, appear to be the
most frequent entries appearing in the maintenance logbook.
More than 50 relatively major failures associated with these
system components were documented to have occurred during the
system lifetime.  Further, these Rayco system failures are
increasing with system age.  The failures rest predominantly
with the inability of the Rayco cylinder to open the fume
main doors and thereby actuate the system to collect coke-
pushing emissions.  Thermal overload protectors in the cylin-
der effectively prevent the system from reaching full opera-
tional mode when the doors cannot be opened fully.  The con-
tributing causes include:  bent doors, degradation of the
hood snorkel lining which causes the fume main doors to bind,
excessive suction in the fume main, and alignment problems.
Most of these problems tend to increase with age.  This is
especially true of the relatively complex fume main system
that must be accessed by the collection hood during each and
every push.  The harsh environment above the coke oven bat-
tery certainly contributes to gradual degradation of this
portion of the emission control system.

From those malfunctions observed when the on-site Clayton ob-
servers evaluated the system visually, it appeared that the
control system operator (also the coke guide operator) was
frequently unaware of control system failure.  In one such
instance, it was observed that the Rayco cylinder tripped
the circuit breaker after traversing approximately 90 per-
cent of its stroke in attempting to open the fume main doors.
The limit switch, due to the system design, did not activate
and open the louvers on the main fan.  Therefore, coke-push-
ing emissions were not captured or scrubbed.

Failures associated with controls and instruments constitute
the fifth category of system downtime.  About 10 failures
due to these components were noted throughout the three-year
system operation.  These failures, therefore, are not a major
contributor to system downtime; significantly,  the majority
of these failures occurred during the initial year of system
operation.

The historical performance, in general, indicates that between
150 and 200 failures of the system, corresponding to between
10 and 20 percent of system operating time, occurred during
the control system's three-year operation as documented in
the maintenance logbooks.   This estimate is conservative be-
cause the noted downtime did not always involve entire days
of system shutdown.

Figure 6.1.1 displays trends in maintenance requirements with
regard to historical economics as deduced from cost accounting
records at Ford Motor Company.  These data indicate that ex-
cessive maintenance expenditures or trends in system degrada-
tion (manifest by gradual increase in maintenance cost

-------
                             - 85 -
     requirements) cannot be attributed to this control system.
     Accurate assessments of man-hour requirements or part replace-
     ment costs could not be deduced from the limited cost data
     available.  The costs of maintenance of the Ford/Koppers emis-
     sion control system amount to approximately $0.15 per ton of
     coke produced, on a yearly basis.


8.2  Assessment of the Abatement Efficiency

     Results presented in Section 7.0 indicate that,  when opera-
     tional, the control system abates  pushing emissions with an
     efficiency ranging from nil to 65  percent, depending upon
     the contaminant considered.  While scrubber efficiencies are
     relatively high for most contaminants, the ability of the
     hood to capture emissions under a  wide variety of push condi-
     tions is seriously questioned.  As discussed previously,
     emissions must be captured and scrubbed to be considered
     abated.  Therefore, total abatement efficiency is decreased in
     almost direct proportion to hood capture efficiency.  Tables
     7.0-1 to 7.0-14 indicate that, in  general, the system is
     capable of capturing up to 65 percent of the pushing emis-
     sions (depending upon greenness).   The scrubbing efficiency
     for filterable particulate is greater than 99 percent.   Since
     the predominant pushing emissions  (on a mass basis) consist
     of particulate matter,  we conclude that the Ford/Koppers con-
     trol system reduces the most significant portion of the coke-
     pushing emissions to between one-half and one-fifth of their
     uncontrolled mass emission rate when only a moderate number
     of green pushes are encountered.  Considering the EPA esti-
     mate of 30,000 tons of  particulate and gaseous contaminants
     annually emitted from all coke ovens'^^, between 15,000 and
     24,000  tons of particulate could be prevented from entering
     the atmosphere annually if all coke ovens in the U.S.A. were
     equipped with this control system.  Therefore, this emission
     control system is certainly a contribution to the developing
     abatement technology required to control this large source  of
     atmospheric contaminants.   However, certain process and sys-
     tem modifications are required to  upgrade the overall control
     (capture and scrubbing) efficiency to a higher level.

     These conclusions are contingent upon the accuracy of fugitive
     emissions estimating techniques used in this study.


     8.2.1  Hood Performance

            Sections 6.2 and 6.3 discussed the method of determin-
            ing the hood capture efficiency.  Regression analyses
            indicate that Method 1 provides a correlation which,
            statistically, is more highly significant (-0.99) than
            that defined by  Method 2.  In Method 2, the engineer-
            observer's visual estimate  of hood capture efficiency
            was correlated to the average greenness rating of the

-------
                         - 86  -
       coke push.  This correlation exhibited  a correlation
       coefficient of -0.90.  This indicates that  the esti-
       mate of average hood capture efficiency by  visual
       techniques is well correlated with the  coke push
       greenness rating.  However, this empirical  correla-
       tion does not necessarily indicate a high accuracy of
       the absolute value of the estimate of hood  capture
       efficiency.  It is tempting to conclude that the
       second method provides a better correlation.  Yet,
       almost certainly, the engineer-observer is  somewhat
       prejudiced by the greenness of the push in  trying to
       determine the hood capture efficiency visually.  The
       extent to which the estimate is numerically erroneous
       is indeterminable.  A useful guide is available from
       the Method 1 results which are more rigorous in nature
       but suffer from some variations in the  experimental
       and process conditions.  Without additional data or
       more refined techniques for estimating  fugitive emis-
       sions, we conclude that these two estimating tech-
       niques should provide upper and lower bound estimates
       of hood capture efficiency.

       It should be emphasized that the data obtained are
       firmly tied to process conditions existing during hood
       efficiency testing.  If some anomalous  process factors
       produced atypical push conditions, the  results are
       biased accordingly.

       To assist in depicting the degree to which the hood
       design contributes to emissions capture depending upon
       push greenness, Figures 8.2.1-1 through 8.2.1-3 dis-
       play photographically the fugitive hood emissions for
       clean, moderate,  and green pushes, respectively.  Two
       views are provided for each type of push.  For each
       view, three photographs show the approximate beginning,
       middle, and termination of the push.  The represented
       photographs are chosen from a large selection and are
       judged typical of each type of push.

       These photographs visually depict the extent of fugi-
       tive emissions which when estimated by the best avail-
       able techniques indicate that hood design is likely
       inadequate.  Table 8.2.1 documents oven numbers and
       push times associated with the photographs.
8.2.2  Scrubber Perfo rmanc e

       Clayton Environmental Consultants concludes that the
       Kinpactor scrubber is performing well in abating par-
       ticulate pushing emissions considering the relatively
       high control efficiency which the scrubber exhibits
       for filterable particulate emissions.  The documented
       control efficiency is somewhat lower than that reported

-------
                 . 87 _
in the Design Manual; however, the scrubber pressure
drop measured during this evaluation program was some-
what less than design conditions.

According to Table 7.0-5, scrubbing efficiencies for
sulfur dioxide average 42 percent.  Normally, liquid
scrubbing of sulfur dioxide is more- efficient.  In
this application, however, the concentration of sul-
fur dioxide in the inlet gas is very low.  Scrubbers
used to abate sulfur dioxide emissions (in the range
of 500 to 3,(f)00 parts per million of sulfur dioxide)
adequately control gas streams.  However, this is
contrasted to the concentrations found in this study,
averaging slightly more than 10 parts per million.
The second contributing factor causing a low sulfur
dioxide scrubbing efficiency is likely attributed to
the recirculated scrubbing liquor.  As the scrubbing
liquor becomes saturated with sulfur dioxide, the
ability o£ the water to collect and retain sulfur
dioxide is somewhat impaired if additive chemical
neutralizers are not used in the scrubbing liquor.

The evaluation of the collectability of sulfur triox-
ide in the Kinpactor is limited by the data available.
Clearly, the data in Table 7.0-6 are sparse and widely
scattered.  These data do not permit meaningful con-
clusions due to the extremely low concentrations of
sulfur trioxide detected.  Despite the two low effi-
ciency points cited in Table 7.0-6, we expect that the
Kinpactor removes sulfur trioxide emissions relatively
efficiently from the incoming airstream.  High removal
efficiencies are found frequently in scrubbers of much
lower contact power.  Consequently, it may be expected
that the Kinpactor reduces what little sulfur trioxide
is emitted and captured during a push to an acceptable
outlet level.

The ability of the Kinpactor to remove carbon monoxide
emissions to the level indicated in Table 7.0-8 is
surprising because no additives are introduced into
the recirculated scrubbing; liquor and because of the
low solubility of carbon monoxide in water.  Generally,
even high-energy scrubbers are incapable of removing
carbon monoxide.  However, the data indicate that in
at least one test, control efficiencies in excess of
80 percent were measured.  Again, however, the low
concentrations of carbon monoxide found contribute
some uncertainty to the measured control efficiencies.

Table 7.0-9 shows that the scrubber efficiency for
oxides of nitrogen is zero in all 13 tests.  A sta-
tistical analysis of the inlet and outlet concentra-
tions was undertaken to determine whether the inlet
and outlet measured concentrations are actually

-------
                  -  88  -
distinguishable in a statistical  sense.   The  results
indicated that, due to the variability  in measurement
and the scatter of the data,  the  mean outlet  concen-
tration was indistinguishable  from  the  mean inlet
concentration.  As such, control  efficiencies were
concluded to be nil.

As indicated in Section 5.2.4,  the  technique  for sam-
pling and analyzing for oxides  of nitrogen is an EPA
reference method and involves  grab  sampling.  An evac-
uated flask is used to sample  a certain amount of the
gas stream at the inlet or outlet during  a 15-second
interval.  Timing is critical  in  that the  inlet and
outlet sampling stations must withdraw  samples from
the stack gas simultaneously while  taking  into ac-
count the time lag between the  inlet and  outlet loca-
tions.  Attempts to refine this method  produced no
significant improvement in resolving differences in
concentrations from inlet to outlet.  It  can be in-
ferred, therefore, that low scrubber efficiencies for
oxides of nitrogen are attributable either to a large
variability in oxides of nitrogen concentrations dur-
ing a push or that the scrubber does not  measurably
remove oxides of nitrogen.  The solubility of nitro-
gen dioxide, usually the principal  constituent of
oxides of nitrogen emissions,  is  such that measurable
amounts should be dissolved in  the  scrubber water.
However, because the scrubbing  liquor is  recirculated,
it is possible to saturate the  scrubbing  liquor in a
short period of time.  Therefore, it is likely that if
the scrubber does exhibit a measurable  abatement effi-
ciency for oxides of nitrogen,  this efficiency would
be quite low.

Tables 7.0-10 and 7.0-11 indicate a high  variation in
scrubber efficiency for benzene emissions as well as
benzene and homologue emissions,  respectively.  High
molecular weight benzene homologues are, of course,
more easily scrubbed because these  compounds tend to
exhibit lower volatility.   The  exact make-up of tKe
mixture of benzene homologues cannot be determined
easily.  Therefore, no judgments  can be made as to
the average molecular weight of this group of con-
taminants.  However,  the high variability in calcu-
lated control efficiency attests  to either a wide
variation in the species being emitted to the scrub-
ber or- simply a scatter in the measured concentra-
tions.  In any event,  data do tend  to indicate that
the scrubber efficiency can range from zero to 94
percent for these types of emissions.  Aqueous scrub-
bing of organic compounds,  especially using recircu-
lated scrubbing liquor, can be expected to yeild low
control efficiencies  in any aqueous scrubber.   The
data in this case do  tend  to indicate that the

-------
                 - 89 -
Kinpactor is removing measurable amounts, to an average
of 34 percent of benzene and benzene homologues emis-
sions .

Total light hydrocarbons represent a broad classifica-
tion of organic compounds whose molecular weight can
range from 16 a.m.u. (atomic mass units) on up.  The
ability of the scrubber to remove these compounds,
especially using an aqueous scrubbing liquor, depends
to some extent on the molecular weight as it deter-
mines approximate volatility or approximate condensa-
tion temperature.  As indicated above, variations
either in measured inlet or outlet concentrations due
to natural data scatter or variations in pushing emis-
sions can account for control efficiencies which range
from zero to 98 percent for total light hydrocarbons
as shown in Table 7.0-12.  However, four of the six
data points do indicate that the scrubber produces
immeasurable reductions in total light hydrocarbons.

Similar scatter of data is evident in Tables 7.0-13
and 7.0-14.  This scatter is attributable to a varia-
tion in pushing emissions or to inherent measurement
errors.  In any event,  the ability of the Kinpactor
to remove methane and homologues, and ethylene and
homologues is not quantified firmly enough to permit
conclusions as to the characteristic scrubbing effi-
ciency of the Kinpactor in abating these substances.
The data do indicate, however, that the scrubbing ef-
ficiency is likely to be low and between zero and 40
percent.

As indicated, sampling measurements were undertaken
for the other species listed in Table 3.2-2.  Because
all values indicated existing concentrations which are
less than the limit of detectability for the analytical
technique required, no accurate assessment of scrubbing
efficiency could be made.  In general, however, it is
likely that high molecular weight substances are re-
moved to a measurable if not efficient degree in the
Kinpactor scrubber.  This conclusion is based upon the
other non-zero scrubbing efficiencies documented in
Tables 7.0-1 through 7.0-14.  In general, those species
which are particulate in nature are likely removed to
an efficiency of 90 percent or greater.  Gaseous con-
taminants are likely removed to an efficiency somewhere
between zero and. 40 percent, depending upon the molec-
ular weight of the contaminant and the solubility in
the recirculated aqueous scrubbing liquor.

-------
                             - 90 -


8.3  Applicability to Other Installations

     The question of whether the scrubber control efficiencies,
     hood capture efficiencies, and total system abatement effi-
     ciencies are those which can be expected to be encountered
     in other coke batteries is related to many factors.  In
     general, the geometry and operation of the coke ovens at
     Ford Motor Company are nearly identical to other installa-
     tions.  One process-related parameter, the degree-of-green-
     ness, is an important consideration in extrapolating ex-
     pected system performance at other batteries.   At installa-
     tions where the frequency of green pushes is less than that
     encountered during the testing period at the Ford Motor
     Company installation, it is expected that hood capture effi-
     ciencies would be measurably improved over those reported in
     this study.  Conversely, coke facilities which experience
     more green pushes should expect increased fugitive (uncap-
     tured) emissions from a similar control system over those
     reported in this study.

     The question of representativeness must be addressed as it
     relates to the average push greenness encountered at Ford
     during the test period.  Using the engineer-observer's esti-
     mate of the degree-of-greenness (having concluded empirically
     that the Ford observer was biased; see Section 6.3),  the per-
     centage of green pushes experienced during testing is com-
     pared with the percentage of green pushes documented during
     all other times when the engineer-observer was on-site.   Re-
     sults show that, during the  testing documented in Tables
     7,0-1 to 7.0-14, only 47 percent of the pushes could be clas-
     sified as "clean" during the test period, whereas 64 percent
     were "clean" pushes  during non-test periods.   Because hood
     capture efficiency,  and therefore total abatement efficiency,
     is a sensitive function of degree-of-greenness (see Figure
     6.3-4), the specific data in this evaluation may be somewhat
     biased toward a detrimental  evaluation.   The extent to which
     47 percent clean pushes is characteristic of other coke-making
     facilities determines the  degree to which expected control ef-
     ficiencies can be extrapolated  to those facilities.

     While undoubtedly the geometry  of ovens (mainly the height of
     the ovens) influences most directly the hood capture  effi-
     ciency, collection efficiencies can be obtained which are at
     least as good as those documented in this study by tailoring
     the hood design to a particular battery being  controlled.

     Unless suitable design modifications are made  to the  collec-
     tion hood-quench car design,  we would expect that collection
     efficiencies would be identical to those displayed in Figure
     6.3-4, assuming similar push greenness.   The best collection
     efficiency estimated (subjectively),  even for  a clean push,
     is approximately 90  percent.   By measurement,  the collection
     efficiency is substantially  less than this,  being about  60
     percent for a clean  push.   From the system observations,

-------
                             - 91 .
     these low efficiencies are likely attributable to both the
     system design and the changes made to the system throughout
     the life of the control system.  We emphasize that these con-
     clusions are based upon the accuracy and precision of the
     techniques used to estimate fugitive emissions.

     In summary, we conclude that the data obtained in this study
     are representative of the Ford/Koppers control system per-
     formance.  Therefore, these data are considered applicable
     to predicting total abatement efficiencies at  other  installa-
     tions.   However,  estimated hood capture  efficiencies  set
     forth in  this report  should be  interpreted with an understand-
     ing  that  the percentage of green pushes  during the sampling
     phase of  the study was somewhat atypical of routine  process
     conditions at the Ford coking  facility.

8.4  Limitations of the Study

     The evaluation of the Ford/Koppers control system was de-
     signed to provide data and quantitative estimates useful in
     assessing the long-term performance of the system with re-
     gard to  maintenance,  operability,  and the ability of the sys-
     tem to abate pushing  emissions effectively.   These,objectives
     are fulfilled by the  results presented in this report.  How-
     ever, a  few considerations that are relative to the  applica-
     tion of  this sytem for controlling coke-pushing emissions
     have not been addressed in this project.

     The original roster of contaminants displayed in Table 3.2-2
     comprises those substances Clayton Environmental Consultants,
     Inc. expected to find in the coke-pushing emissions.   Many
     of these were not present at the lower limits of analytical
     detectability.   Therefore, if some compounds in the  push emis-
     sions have gone undetected in this study, it is because of
     analytical sensitivity or because  their potential existence
     was largely unknown when the study was designed and  executed.
     In light of its overall experience with coke oven emissions,
     Clayton  Environmental Consultants, Inc. believes that the
     roster of contaminants sampled in  this study is unusually
     comprehensive and representative of the substances believed
     to be present in coke-pushing emissions.  Clearly, however,
     the nature and concentration of these species could  vary
     from installation to  installation  depending  upon the  particu-
     lar type of coal used and the coking practices employed at
     the facility.

     The possibility for a diversity of chemical  reactions between
     the scrubbing liquor  and its contents and the contaminants
     coming into the scrubber gives rise to legitimate questions
     regarding by-product  or secondary  emissions  from the  scrub-
     ber which could go largely undetected.  The  chemistry of
     reaction between recirculated scrubbing liquor and high
     molecular weight contaminants, while being unlikely,  could
     produce  chemical species whose physical and  chemical  properties

-------
                        - 92 -
do not lend themselves to collection or analysis by the sam-
pling and analytical procedures used in this study.  Further-
more, any accumulation of collected emissions in the recir-
culated scrubbing liquor, and chemical reactions among these
captured contaminants, could lead to emission to the atmos-
phere of other materials than those sampled.  This study
neither presumes to confirm nor negate such possibilities.

-------
                             _ 93 _
                      9.0  RECOMMENDATIONS
Because this comprehensive evaluation indicates that the perform-
ance of the Ford/Koppers system could be improved significantly—
considering both hood capture and scrubber efficiencies —• some
improvements in the design and operation should-be made to upgrade
performance.  Our recommendations include alterations to the hood
design, quench car design, scrubber design, fan capacity, and
other general aspects of the control system.


9.1  Hood-Quench Car Clearances

     The post-start-up modifications delineated in the Koppers
     Design Manual have contributed to some decrement in hood
     capture efficiency compared with the original design.  By
     increasing the gap between the hood sides and the built-up
     quench car, air infiltration and consequent contaminant es-
     cape have increased.   Furthermore, by removing the synchro-
     nization unit and thereby requiring that the end of the
     quench car be removed for the operator,  fugitive emissions
     have been further increased.  All of these changes have
     contributed to a decrement in hood capture efficiency esti-
     mated to be in the order of 20 to 30 percent below the orig-
     inal design specifications.  While no recommendations can be
     made as to how to eliminate the problem of mobile side
     plates on the hood, as they degrade in this environment, and
     the necessity for the operator to see the coke, it is cer-
     tain that this air gap must be minimized if hood capture ef-
     ficiency is to be improved*


9.2  Redesign of Hood Structure

     Aside from building up the sides of the  hood to reduce clear-
     ances, capture efficiency could be improved by redesign of
     the hood structure.  Because the active  portion of the hood
     covers less than 50 percent of the quench car at any one
     time (see Section 4.2), a considerable portion of the hood
     face presented to the quench car is made up of a solid plate
     designed to conduct emissions to the center of the hood.  If
     the entire hood were  made an active collector (that is, by
     making the entire hood cross section an  open area), undoubt-
     edly a large percentage of fugitive emissions would be col-
     lected.  However, because it is necessary for the quench car
     to move underneath the hood to provide a level load of coke,
     green pushes will continue to emit contaminants from that
     portion of the quench car not under the  hood.  The only ap-
     parent solution is to provide a hood large enough to cover
     the entire quench car, including the length that the quench
     car must move leveling the coke.  Such an alteration would
     require a large increase in exhaust ventilation volume.
     This is especially critical to operating costs in considera-
     tion of the large pressure drop already  required in the

-------
                             - 94 -
     venturi scrubber.  Nonetheless, it is likely that' the only
     practical way to effect a measurable increase in the hood
     collection efficiency is by redesign of the hood and by
     increase in the hood ventilation exhaust volume.
9.3  Scrubber Operation

     As indicated in Section 8.2.2, the scrubber is very effi-
     cient in abating particulate matter conveyed to it.  The
     extent to which this collection efficiency can be increased
     is largely dependent upon the ability of the system to be
     made to operate with a larger pressure drop across the
     throat of the scrubber.  Currently, liquid-to-gas ratios
     are sufficient and are in agreement with the design concept.
     As such, the only room for improvement likely rests in in-
     creasing turbulent contact in the scrubber throat.  Sub-
     stantial removal of other contaminants can likely not be im-
     proved upon without the use of chemical additives to the
     scrubbing liquor.  However, the merits of doing so are ques-
     tionable because of the already low concentrations of these
     contaminants measured in the sampling phase of this study.


9.4  Process and System Modifications

     Opacity levels from the scrubber outlet stack are unneces-
     sarily high for the green pushes.  Unless green pushes can
     be eliminated, it would be necessary to upgrade scrubber
     performance in order to comply with existing opacity regu-
     lations.  This can be done largely by providing more fan
     power, if required, closing off the venturi throat to pro-
     vide intense turbulence,  and, generally, doing whatever is
     necessary to minimize the number of green pushes on the bat-
     tery.

     As indicated in the discussion of maintenance problems, the
     movable and otherwise somewhat complex mechanisms subjected
     to the harsh environment  adjacent to the coke side of the
     batteries are subject to  gradual deterioration, and there-
     fore induce chronic maintenance problems.  Failure of the
     Rayco  cylinder to fully open the doors of the fume main of
     each oven can likely be corrected by installation of a
     larger, more powerful unit.  Furthermore, redesign of louvers
     and doors on the fume main will minimize and eliminate door
     binding problems and bent levers.  Additionally, by making
     use of the originally designed butterfly damper, excessive
     suction in the fume main  could be eliminated during the
     time when the Rayco cylinder is attempting to open the doors.
     These  modifications, while relatively simple, should provide
     a  system that requires less maintenance and consequently
     higher reliability in abating pushing emissions a larger
     percentage of the time.

-------
                        - 95 -
As indicated, the design of the automated, main fan louvers
requires that the fan change from an idle to fully-operational
mode once the fume main doors are opened completely.  Unfor-
tunately, based on long-term system observation, the coke
guide operator sometimes actuates the system 15 minutes prior
to a push (depending upon production scheduling between "A"
and "B" Batteries).  Because this results in excessive and
unnecessary energy consumption, the installation of an al-
ternate control scheme for the main fan louvers is recom-
mended.  By using the pusher ram movement as a signal to the
emission control system, the fan would not assume full opera-
tion until the beginning of the push.  A timer circuit would
be useful to provide the proper shutdown sequence after the
pusher ram withdraws from the oven just pushed.

Lastly, by minimizing green pushes, the control system would
function at a much higher level of emissions abatement.  Not
only would overall efficiency be improved measurably, but
the absolute amount of emissions which are being controlled
to a given efficiency level would result in a further in-
crease in absolute abatement.  Certainly, maintenance per-
formed at Ford is as thorough as that performed at other
coke batteries.  However, the abnormally large number of
green pushes observed during portions of the June and July
testing indicated that some malfunctions in process condi-
tions are possibly responsible.  This would indicate that
as much or more attention must be given to minimizing emis-
sions by manipulation of the process conditions as is given
to installation and operation of pushing emission control
systems.  Such process modifications might include extension
of coking time or increased energy input to the ovens to
raise coking temperature.  Only then can the best possible
performance be attained from application of such a coke-
pushing emission control system.

-------
                        - 96 -
                      REFERENCES
Annual Statistical Report, American Iron and  Steel  Institute,
1974, B.O.M. Statistics, and ACCCI Estimates.

U.S. Environmental Protection Agency,  C omp i1a t ion of  A i r
Pollution Emission Factors, 2nd edition, April,  1973.

Camp, J.M. and C.B. Francis, The Making, Shaping, and Treating
of Steel, 5th edition, Carnegie-Illinois Steel  Corporation,
Pittsburgh, Pennsylvania, 1960.

Air Pollution by Coking Plants, United Nations  Report - Economic
Commission for Europe, ST/ECE/Coal/26, 1968.

"Standards of Performance for New Stationary  Sources," Federal
Register, 40CFR60, June 14, 1974.

Paley, L.R., et al., "Were the Measured Emission Rates Repre-
sentative of a Coke Battery's Typical  Emissions?" presented
at the Source Evaluation Society, East Central  Air  Pollution
Control Association meeting, Dayton, Ohio,  September  19,  1975.

Semrau, K.T., "Correlation of Dust Scrubber Efficiency,"  Journal
of the Aij^ Pollution Control Association, Volume 10,  No.  3,
June,1960Y

Bowker, A., and G.J.Lieberman , Engineering  Statistics, Prentice-
Hall, Inc., 1959.

-------
                             - 97 -
                 FOREWORD TO FIGURES AND TABLES




     The following figures and tables are numbered according to the




report section in which they are first presented.  In some cases,




"families" of figures and tables bear the same section number; mul-




tiple figures and tables within a section are appropriately suffixed




by a hyphen followed by a sequential number.  Some families of tables




and/or figures, however, are interrupted by figures and/or tables hav-




ing the same section number depending upon the order of presentation




in the report text within a given section.

-------
                        - 98 -
                       TABLE  1.2

 CHARGE RATE AND EMISSION FACTOR CONVERSION EQUATIONS

                  Ford Motor Company
              Rouge Plant - Coke Battery
                  Dearborn, Michigan
Charge Rates;
     Ton Dry Coal Charged               Ton Wet Coal Charged
             Hour	  =  °'938  X  	"	
Ton Coke Produced
                                     Ton Wet Coal Charged
                                                 1
Emission Factors:
  Pound  Particulate
Ton Dry Coal Charged
                                          Pound  Particulate
                                       We(.
                                                     charged
     Pound Particulate                Pound Particulate
     Ton Coke Produced  s  1.342  X  ton Wet Coal Charged
 Based on Ford laboratory report of coal moisture content
 and typical coke yield for June and July, 1975.
                         Clayton Environmental Consultants, Inc

-------
                                               -  99   -
                                            TABLE 3.2.1

                  EMISSION  FACTORS FOR  UNCONTROLLED METALLURGICAL
                                        COKE MANUFACTURE

              (From  EPA  Compilation of  Emission Factors,  April,  1973)
                 EMISSION FACTORS FOR METALLURGICAL COKE MANUFACTURE WITHOUT CONTROLS*
                                      EMISSION FACTOR RATING: C


Type of operation
By-product coking0
Unloading
Charging
Coking cycle
Discharging
Quenching
Underfiring*
Beehive ovens*

Partbulates
Ib/ton

0.4
1.6
0.1
0.6
0.9
—
200
kg/MT

0.2
0.75
0.05
0.3
0.45
—
100
Sulfur
dioxide
Ib/ton

—
0.02
—
—
—
4
—
kg/MT

—
0.01
_
—
—
2
—
Carbon
monoxide
Ib/ton

—
0.6
0.6
0.07
—
_
1
kg/MT

—
0.3
0.3
0.035
—
—
0.5

Hydrocarbons"
Ib/ton'
•

—
2.5
1.5
0.2
—
—
8
kg/MT

—
1.25
0.75
0.1
—
—
4
Nitrogen
oxides (N02)
Ib/ton

—
0.03
0.01
—
—
—
—
TtgTMT

—
0.015
0.006
—
—
—
—

Ammonia •
Ib/ton

—
0.02
0.06
0.1
—
_
2
kg/MT

—
0.01
0.03
0.05
_
—
1
*EmWon factor* expressed as units per unit weight of coal Charged.
''Expressed n methane.
'References 2 and 3.
^Reference S. The culfur dioxide factor is bond on the following representative conditions: (1) sulfur content of coil charged to oven is 0.8
percent by weight; (2) abopt 33 percent by weight of total sulfur in the coal charged to oven is transferred to the coke-oven gat; (3) about 40
percent of coke-oven gat is burned during the undarf iring operation and the remainder is used in other parts of the steel operation where the rest of
the sulfur dioxide is diicharged-about 6 Ib/ton (3 kg/MTI of coal charged; and (4) gas used in undarfiring has not been desulfurized.
•R*farancas1and4.

-------
                           - 100 -
                         TABLE  3.2-2

                   SUMMARY  OF CONTAMINANTS
                EXPECTED  IN PUSHING EMISSIONS

                     Ford Motor Company
                 Rouge Plant -  Coke Battery
                     Dearborn,  Michigan
Contaminant
Acetylene
Acridine
Acrolein
Ammonia
Anthracene
Benzene & homologues (as CgHg)
Benzo(a+e)pyrene
Carbazole
Carbon disulfide
Carbon monoxide
Chlorine
Chrysene
Cyanide
Ethylene & homologues (as C2H^)
Fluoranthene
Fluorene
9-Fluorenone
Hydrogen sulfide
Methane & homologues (as CH^)
Naphthalene
Nitrogen oxides (as N02)
Particle sizing
Particulate
Phenanthrene
Pyrene
Pyridine
Sulfur dioxide (as S02)
Sulfur trioxide (as 803)
Total light hydrocarbons (as CH4)
Total phenolics (as Cg^OH)
Total reduced sulfur (as S=2)
Total sulfur oxides (as 804)
Detectable
At Inlet*
No
No
No
No
No
Yes
	
-No
No
Yes
No
No
Yes
No
No
No
No
	
Yes
No
Yes
	
— : —
No
No
No
Yes
No
Yes
Yes
—
— —
Included In
Further
Testing
No
No
No
No
No
Yes
Yes
NO
Yes
Yes
No
No
No
Yes
No
No
No
No
Yes
No
Yes
Yes
Yes
No
No
No
Yes
Yes
Yes
No
Yes
Yes
* According to "range finding" tests conducted during January,
  February, March, or April, 1975 wherein at least one sample
  showing a concentration above the limit of detectability
  classifies the material as detectable.
                         Clayton Environmental Consultants, Inc.

-------
                      FIGURE 5*1.1-1
SPECIMEN OF RECORD FOR OPERATING CHECKS BY BATTERY FOREMAN
            ONCE EACH SHIFT FOR CONTROL SYSTEM
                    Ford Motor Company
               Rouge Plant - Coke Battery
                    Dearborn, Michigan

ITE
HFT


FAR BEARING TEMP. ABOARD .
"A" STACK ROOM OUTBOARD
MOTOR BEARING TEMP. INBOARD
"A" STACK ROOM OUTBOARD
FAN AMPERAGE
•HEATER ROOM
CAS TEMP. j
RECORDER ]
"A"&nB" <
HEATER ROOM i
KINPACTOR
DIFFERENTIAL
"A"&"B" ATE. BEFORE PUSH
AVE. DURING PUSH
ft BLACK FLUME MAIN
•2 TOD KINPACTOR INLET
tt GREEN FAN INLET
fk BLUE STACK
BETWEEN PUSHES
DURING PUSH
WATER FLOW FLOW CHART I A&B HEAT, ROOM)
TO PRESS. AT PUMP (A STACK)
KINPACTOR PUMP OPERATINGfN. OR S.)
DRAIN VALVE ONE TURN (A STACK)
OBSERVE ONE PUSH FROM STACK FLUME
AND RECORD CONDITION BI NOON
fRTNGT.EMANN SMOKE CHART)
NOTES
OPERATING CHECKS ON SI
/ /
1-B>
4*3
13,5"
1^,5"
> 40
t.20
I JO
t'

+f



/fIG
•4ft 	
/5*r
x^rr
/Tf-'
/,?*•
//«•'
3V-U

M/C
/K.C We-
tejLeit
j

tJt<-'
&
A/



///o
*z&
|XL5
ISJS'
y.<-t> _
/Ai
// 0
£L*40
t
(
(
\
]
t
1
vTXf
A/



*)KFTrR-<:is ^
BY BAT1
f-/y
4**5^

/'*
/3V
S0&
//o
XfO
30
•30
3^
"75"
-««u\
J
73.Q
&(•
A/

0

TJSHING SI,
ERT FOREM
/-/*
ff ?
A
'fLfL
QtPO*'
S9*fCJL

A
H
/l^tKet









f
STEM ONCE EACH 8 HR. SHIFT
UN
f/ty
*+/
/«*•
1 /•?=*•
/'VA
/*-
—
—
•^
—
--
—
—


rV
S


•#'&t*A
t-f,9tr*
//*£'>
a+T
Ft&A -
/r*7>^^
///*
ar/
/,V.<"
/3.-
ftvt -j?«f«r
—
—
—
^•M
—


*•/**
^


72vsS.'A
t>*$«"
.Cte^S 71
1&&C4P
I-IL
* *>
i ^^T
tJfey.
i :ij?"
/ /s-
c-t«*-#.

/
(



                                                                                      I
                                                                                      f->
                                                                                      o
                                                                                      l-»
                                                                                      I

-------
                    -  102 -
                FIGURE 5.1.1-2
SPECIMEN OF RECORD FOR A-DOOR MACHINE  DOWNTIME
              Ford Motor Company
          Rouge Plant - Coke Battery
              Dearborn, Michigan

                                       tfav*. Atett.9.
                                       . #- '..' •*     y '*  ,' f .| r* . t»
                        ««. . . •«

-------

—V
**'fjihf

-------
                              -  104  -

                         FIGURE 5.1.1-4

 SPECIMEN OF  DATA EXTRACTED FROM MAINTENANCE DEPARTMENT LOGBOOK

                       Ford Motor Company
                   Rouge Plant - Coke Battery
                       Dearborn, Michigan



7-22-74     Put A-battery pollution fan motor back In service at

            6:10 PM.   A-door machine out of service from 8:00 AM

            to 10:10  AM to change  fume hood travel drive wheel.



8-29-74     Flume damper handle at oven 32, A-battery needs

            straightening.



8-31-74     A-battery fume main door arms (on oven) numbers 39,

            41, 43, 48, and 49 need to be straightened.



9-1-74      We straightened the fume main door arm at 39.



9-30-74     A-battery pollution fan is down for repair for two

            days.



10-18-74    Angle clips were installed...on A-battery fume main.

            South beam was slightly twisted at top so we had to

            drive wedges...gave machine to operations at 4:45 PM.

            A-battery emissions control fan in service at 5:00 PM.

            All covers were installed on Kinpactor.



11-5-74     A-door machine fume hood is not operating all the time.

            We found  that the hood carriage doesn't move far enough

            to trip LSDO located on the south side of the Rayco

            platform.  The door is opened and the hood opening is

-------
                              -  105  -

                         FIGURE 5.1.1-5

  SPECIMEN OF DATA EXTRACTED FROM OPERATIONS DEPARTMENT LOGBOOK

                       Ford Motor Company
                   Rouge Plant - Coke Battery
                       Dearborn, Michigan


6-26-75     Greased  the rollers  on the Rayco unit.  A-door machine

            out of service  9:40  to 10:15 maintenance making adjust-

            ments to overload  limit  for Rayco flume hood.




6-27-75     A-door machine  out of service from 8:00 to 8:30

            maintenance working  on flume collector, Rayco, and hood.

            A-door machine  out of service 12:40 to 1:45 working on

            Rayco again.



7-1-75      A-door machine  out of service 11:30 to 2:00 working on

            Rayco cylinder.



7-3-75      Fume collector  down  at 4:37 PM due to substation 46A

            going out.   Back on  at 9:00 PM.



7-7-75      A-door machine  out of service 11:45 to 1:00, maintenance

            replaced flume  hood  extension.



7-8-75      A-door machine  out from beginning of shift through 9:15;

            maintenance worked on flume collector hood slide.



7-11-75     Fan motor tripped  out 2:45 AM;  back on for first oven

            after lunch at  4:10  AM but it's not working very well.

            The extension is going against the main also it is

            tripping the  Rayco reset...fan motor tripped out again

            at  6:15.

-------
                  DAILY LOG OF FUME HOOD PERFORMANCE
         Ford Motor Company * Rouge Plant * Dearborn, Michigan
Date:
Atmospheric Pressure
Initial:
Push
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15

Avg.
Oven .
No.

















Schedule
Time
0800
0815
0830
0850
0905
0920
0935
0950 •
1005
1020
1035
1055
1110
1125
1145


Push
Time







.









Cross-Draft
Velocity
(fpm)





<^











Fume Hood
Capture
(X)

















Stack
Opacity
(7.)

Coke Side
Condition

FIGURE 5.1.3-1
Net Coal
Charged
(Ibs)


SPECIMEN OF RECORD FOR HOOD
PERFORMANCE
Ford Motor Company
Rouge Plant - Coke Batt
Dearborn, Michigan


























^^^^^^













-------
               Approximately 250'
                                                   Approximately 10'
 Locomotive
Maintenance
  Building
Engineer-Observer
for push on Ovens
46-64
                                                                      Quench
                                                                       Tower
                                   Railroad
                                        ,  Engineer-Observer for
                                    T  "T" push on Ovens 1-44

                                    1 Approximately 10*
                                      Road
                    FIGURE 5.1.3-2

  SCHEMATIC DIAGRAM OF ENGINEER-OBSERVER DURING A PUSH

                  Ford Motor Company
              Rouge Plant - Coke Battery
                  Dearborn, Michigan
                                     Clayton Environmental  Consultants,  Inc,

-------
                 DAILY LOG OF SCRUBBER SYSTEM PERFORMANCE
           Ford Motor Company * Rouge Plant * Dearborn, Michigan
Date:
Atmospheric Pressure:
Initial:
Push














Oven
No.














; "




Push
Time

















Inlet Conditions
Dry
Bulb
(°F)

















Wet
Bulb
(°F)

















Static
Pressure
("H40)

•















Velocity
Pressure
("H20)


















Throat Conditions
Pressure
Drop
("H20)

'











Water
Pressure
(psi)


•










Water
Temp .
<°F)













r FIGURE 5.1.4
SPECIMEN OF RECORD FOR
SCRUBBER OPERATING PARAMETER!
,Ford Motor Company
Rouge Plant - Coke Battery
Dearborn, Michigan
x. ^X
Outlet Conditions
Dry
Bulb
<°F)













\
i

'
Wet
Bulb
<°F)

















Static
Pressure
("11*0)

















Recirculated
Water
Flow
(gpm)

















Fan
Motor
Current
(amp)

















i
h->
0
oo
                                               Clayton  Environmental  Consultants,  Inc.

-------
                                - 109  -
so 8364

-------
                                    110. -
   __ - .._ I»ON MAt>IM« Oft m
 f% SO S427 «• v rrm •«
                   CUM  cmncr.  ton

                  r/.rfr.nv  *
             Ttir
                   r.noss
TARE
— 1
/
                      6

                                                               r
                    7
                                                     y
                                                          £1

                                                  3-
                                                &.
                     ?~L
                                                      X.
                                                     ~L
                                     760
±L
                                                               0
                                                                   a
      Zl
      II
      3L
               l^L
    XXi
                                                                   0
                                                            f
                                     4.60
                                                                   o
            /I
                       FIGURE  5.1.5-2



                   SPECIMEN  OF RECORD FOR

                      COAL CHARGE DATA



                     Ford Motor Company

                 Rouge Plant -  Coke Battery

                     Dearborn, Michigan
                 ^_^^^^^^^^^^^^^^^^^^__^^^^^^^^^^^^^^^^^^^^^^^^^
                                     foreman
                                                                   O

-------
                       -  Ill -


                   FIGURE  5.1.5-3

SPECIMEN OF RECORD FOR COKE OVEN FLUE TEMPERATURES

                Ford Motor Company
            Rouge  Plant -  Coke Battery
                Dearborn,  Michigan
          COKE OVEN FLUE TEMPERATURES


           ^t                  -     OATt _2
 •COCK


 SHIFT
                    NCAOKR
       MCVCMSC  1-4-9
                                 REVERSE 2 - 1
    P.S.
NO.
c.s.
P.S
NO.
c.s.
                                           .-•> '**•
                  I  It
          n
                    n
                       U
                       o
 2  33
          0
          33 ,,
                  2  3,3 0
                 33
                                     15
                       n
                               •n
                                           »'•*'.«
            31
                                           •t-
                                                /)
                                                o
    n
                               ^  o
 ? 00 f)
                              ; 3 ") "
       A
                                K c->
    3 !
                          57
             58
                              5-0*530
                          61
                          62
      * SO 8348
     MAKING OPCR
     AUC 7J

-------
                 FIGURE 5.2

SCRUBBER INLET AND OUTLET SAMPLING LOCATIONS
     . PUSHING EMISSIONS COLLECTION SYSTEM

              Ford Motor Company
          Rouge Plant - Coke Battery
              Dearborn, Michigan
                                                                                 Scrubber Outlet
                                                                               Sampling Locatloi
                                                                 Scrubber  Inlet
                                                                 Sampling  Location

-------
           Hood
                      Hl-Vol
                      Sampler
                                     Ovens
                                                          TOP VIEW
                    Exhaust  Duct
                       2"x4"x9'0'
                                      Standpipes
1
5G
1 VA 1 JT 1 flll
7-
Walkway
and Railii
1
Quench Car —
J
»g ^S
Mobile
Hood ^

"•^
trT\
^

Coke-
Side
Door
Machine

^


1=4
15' «

^
1 Walkway and Railing
_^ Hi-Vol Sampler
A-Battery
/^— Ovens ** -^

2

0'
END VIEW
(Facing north)
FIGURE 5,2.13.1
                                                                                  U)

                                                                                  I
SCHEMATIC DIAGRAM OF HIGH-VOLUME AIR SAMPLING LOCATION

                  Ford Motor Company
              Rouge Plant.- Coke Batt«yy
                  Dearborn, Michigan

-------
                                - 114 -
                             TABLE 6.1.1-1
                 SUMMARY  OF  AVERAGE MONTHLY  UTILITIES

                          Ford  Motor  Company
                      Rouge  Plant -  Coke Battery
                          Dearborn, Michigan
Month
8/73
9/73
10/73
11/73
12/73
1/74
2/74
3/74
4/74
5/74
6/74
7/74
8/74
9/74
10/74
11/74
12/74

1/75
2/75
3/75
4/75
5/75
6/75
7/75
8/75
Overall
Average
6,7,8/75
Average
Main Fan
Current (amps)
Before
Push
129
125
122
114
116
133
131
118
116
119
133
125
121
120
119
119
116

110
170
124
121
145
186
116
116

127

139
During
Push
242
235
235
235
240
233
230
221
223
233
232
239
238
240
239
240
243

242
180
241
235
268
332
214
216

237

254
Kinpactor
Pump
Flow
(gpm)
747
750
680
652
614
MMM
	
— -
948
855
827
852
867
866
799
—
	

880
982
—
954
948
776
839
688

778

768
Pressure
(psig)
45
46
47
48
44
41
45
52
54
44
38
51
43
50
51
49
49

47
52
53
54
57
43
50
40

48

44
Horsepower Used
Fan
Before
Push
1245
1210
1180
1100
1120
1285
1265
1140
112.0
1150
1285
1210
1170
1160
1150
1150
1120

1065
1640
1200
1170
1400
1795
1120
1120

1225

1345
During
Push
2340
2270
2270
2270
2320
2250
2220
2130
2150
2250
2240
2310
2300
2320
2310
2320
2350

2340
1740
2330
2270
2590 ,
3210**
2070
2090

2290

2^60
Pump
23
23
22
21
18
^_
_
—
35
26
21
30
25
29
28
—
—

28
35
_
35
37
23
29
19

27

24
Total
Weighted
Average*
1630
1590
1570
1510
1540
^ ^_ .
— — —
_ _^_
1500
1540
1620
1610
1570
1580
1560
—
	

15 CO
1710
_— _
1570
1830
2290
1470
1460

1610

1740
 *  Assuming  40  minutes  of  idle  fan  and  20 minutes  of  wide  open  fan
   in  each push-hour.
** Suspected erroneous amperage reading
                                  Clayton  Environmental  Consultants,  Inc

-------
                             - 115 -
                         TABLE  6.1.1-2
        SUMMARY  OF  AVERAGE  MONTHLY SCRUBBER PARAMETERS

                      Ford  Motor Company
                 Rouge  Plant -,  Coke Battery
                      Dearborn,  Michigan
Month
8/73
9/73
10/73
11/73
12/73
1/74
2/74
3/74
4/74
5/74
6/74
7/74
8/74
9/74
10/74
11/74
12/74
1/75
2/75
3/75
4/75
5/75
6/75
7/75
8/75
Overall
Average
6,7,8/75
Average
Kinpactor
Differential
(psi)
44
42
40
36
42
31
34
39
43
44
38
(20)
43
41
42
43
42
46
28
43
43
42
46
39
39

40

41
Nozzle
Pressure*
(psig)
27
28
29
30
26
23
27
34
36
26
20
33
25
32
33
31
31
29
34
35
36
39
25
32
22

30

26
Water
Flow
(gpm)
747
750
680
652
614
____
	
___
948
855
827
852
867
866
799
—
	
880
982
—
954
948
776
839
688

778

768
Contact Power
Gas
1 HP ^
\IQQO SCFMJ
6.9
6.6
6.3
5.7
6.6
4.9
5.3
6.1
6.8
6.9
6.0
—
6.8
6.4
6.6
6.8
6.6
7.2
4.4
6.8
6.8
6.6
7.2
6.1
6.1

6.4

6.5
Water
/ HP \
llOOO SCFMJ
\ /
0.15
0.14
0.14
0.14
0.12
____
	
	
0.25
0.16
0.12
__—
0.16
0.20
0.19
• _____
	
0.19
0.24
_____
0.25
0.27
0.14
0.19
0.11

0.18

0.15
Total
/ HP \
^1000 SCFMJ
7.1
6.7
6.4
5.8
6.7
_____
	
____
7.1
7.1
6.1
—
7.0
6.6
6.8
—
	
7.4
4.6
__
7.1
6.9
7.3
6.3
6.2

6.6

6.6
* Assuming 18 psi loss from pump gauge reading due to friction
  and head losses.
                                 Clayton Environmental Consultants, Inc.

-------
               50,000
               40,000
Maintenance Costs
as Year-To-Date
               20,000
               It,000
                                .  .   '   1974

                          	_;	1975
                             F    M
S    ON
      A    M    J    J    A

         Month of the Year

            FIGURE 6.1.1
MAINTENAHCE COSTS OF THE KOPPBRS SYSTEM
          Ford Motor Company
      Rouge 'Plant - Coke Battery
          Dearborn, Michigan
                                                                                   50
                                                                                   40
                                                                                   30
                        Maintenance Cost
                        As Percentage Of
                        Total  Oven
                        Maintenance
                        (*)
                                                                                   20
                                                                                   10

-------
                                  -  117 -

                              TABLE  6.1.1-3

    SUMMARY OF REQUIRED MAIHTBNANCE DOCUMENTED III MAINTENANCE LOGBOOK

                           ford Motor Company
                       Rouge Plant - Coke Battery
                           Dearborn, Michigan
Main Fan &
Fan Louvre*
1972: 7-1 1974:
7-22
7-24
9-4
11-26
11-29
12-21*
1973: 2-11
2-17*
2-18*
2-19*
4-19
4-20*
4-21*
4-22*
4-23*
5-29
6-4*
6-5
6-6*
6-8*
6-19
6-30
7-8
7-14
8-27
8-31
11-26
12-22 1975:
1974: 1-16
1-17
3-22
5-14*














5-18
5-22
5-31
6-2*
6-3*
6-4*
6-5*
6-6*
6-7*
6-8*
6-9*
6-14
6-17*
6-19*
6-20
6-22*
6-24*
7-1*
7-2*
7-7
7-10
7-11*
7-12*
7-13*
7-22
9-30*
10-1*
12-1
1-27
3-7*
6-20
7-11














Pump
1972: 7-30
11-30
12-1
1973: 3-11
3-19
12-14
12-23
1974: 3-22
7-14*
7-15*
7-17*
11-28*
1975: 1-14
2-16
5-21






























Door
Machine
1972: 7-1
10-23
12-2*
1973: 4-29
5-3
5-26
6-4
8-2
1974: 3-27
3-31
8-29
8-31
9-1
10-18































Rayco or
Hood
1972: 7-8*
7-16
7-18
10-25
11-17
11-19
12-17*
1973: 4-3
4-7
4-8
4-24
4-25*
5-9
6-1*
11-1*
11-3
11-4
1974: 1-25
3-22
7-21*
11-5
11-7
1975: 1-9
1-14
1-16
2-5
3-7
3-20
5-2
5-8
5-9
5-10
5-13
5-30
6-13
6-14
6-15
6-21
6-22
6-23
6-24
6-26
6-27
6-29
6-30
Controls &
Instruments
1972: 12-5
12-12
12-28*
12-29*
1973: 6-26
12-22
1974: 1-8
12-12





































* Denotes more than one day down time.

-------
                                        TABLE 6.1.3-1
                                      EMISSIOH TEST SUMMARY
     * Ford Motor  Company--Coke Battery

Stack:    Scrubber Inlet
Material:
Acetylene
Stack Diameter:   66" I'D«
1975
Data
1/10
1/10
1/10
Sampling
Period
Start



AVERAGE

2/4
2/4
2/4
2/4
2/6
2/6
2/6
2/6
2/6
.

14:00
14:45
15:00
15:10
13:10
13:10
13:20
13:20
13:30

Stop















Teat
Ho.
1
2
3


1
2
3
4
6
7
8
9
10

Sample
Weight
(tag)















Sampled
Volume















AVERAGE
Stack
Temp.
CP)
















Stack Gaa
Plovrate
(DSCPM)
98,100
98,100
98,100
98,100

71,900
71,900
71,900
71,900
71,900
71,900
71,900
71,900
71,900

71,900
Peed
Rate*
(tons/hr)















)•
Concen-
tration
(PP»).
<1.5
<1.5
<1.5
<1.5

<12
<12
<12
<12
<12
<12
<12
<12
<12

<12
Emission
Rate
Ib/hr*
<0.60
<0.60
<0.60
<0.60

<3.5
<3.5
<3.5
<3.5
<3.5
<3.5
<3.5
<3.5
<3.5

<3.5
Ib/ton
of feed
















                                                                                                      oo
  * Assuming continuous  pushing
                                                     Clayton Environmental Consultants, Inc.

-------
                                       TABLE 6.1.3-2
                                    EMISSION TEST SUMMARY
      Ford Motor Company--Coke  Battery
Stack:   Scrubber Inlet
                                                Material:
Acetylene
                                               Stack Diameter:  66" 1-D-
1975
Data
6/4
6/4
6/4
6/12
7/16
7/16









Sampling
Period
Start
10:49
13:00
15:19
16:36
10:35
12:03









Stop
10:50
13:01
15:20
16:37
10:36
12:04









Test
No.
1
4
7
10
14
17









AVERAGE
Sample
Height
(mg)















Sampled
Volume
















Stack
Temp.
CF)
















Stack Gas
Flovrate
(DSCFM)
80,300
80,300
80,300
80,300
80,300
80,300









80,300
Feed
Rate*
(tons/hr)
1096
1074
952
1052
1029
1066









1045
Concen-
(ppm)
<0.5
<0.5
<0.5
<0.5
<0.5
<0.5









<0.5
Emission
Rate
Ib/hr*
<0.16
<0.16
<0.16
<0.16
<0.16
<0.16









<0.16
Ib/ton
of feed
<0. 00015
<0. 00015
<0. 00017
<0. 00015
<0. 00016
<0. 00015









:o. 00016
                                                                                                    IO

                                                                                                    I
* Assuming continuous  pushing
                                                   Clayton Environmental Consultants, Inc,

-------
                                         TABLE 6.1.3-3
                                      EMISSIOH TEST SUMMARY
Plant; Ford Motor Company—Coke  Battery

        Scrubber Outlet
Material:
Acetylene
Stack Diameter:
                    90"  i.D.
1975
Date
6/4
6/4
6/4
6/12
7/16
7/16









Sampling
Period
Start
10:49
13:00
15:19
16:36
10:35
12:03









A V E 1 A 6
Stop
10:50
13:01
15:20
16:37
10:36
12:04









B
Test
No.
1
4
7
10
14
17










Sample
Weight
(mg)















Sampled
Volume
















Stack
Temp.
CP)
















Stack Gas
Plowrate
(DSCPM)
86,900
86,900
86,900
86,900
86,900
86,900









86,900
Peed.
Rate*
(tons/hr)
1096
1074
952
1052
1029
1066









1045
Concen-
tration
(ppm)
<0.6
<0.5
<0.5
<0.5
<0.5
<0.5









<0.5
Emission
Rate
Ib/hr*
<0.21
<0.18
<0.18
<0.18
<0.18
<0.18









<0.18
Ib/ton
of feed
<0.00019
<0. 00016
<0. 00019
<0. 00017
<0. 00017
<0. 00017









<0. 00018
                                                                                                      K>
                                                                                                      O
  *  Assuming  continuous pushing
                                                     Clayton Environmental Consultants,  Inc.

-------
                                        TABLE 6.1.3-4
                                     EMISSION ZEST  SUMMARY
Plant: F°rd Motor Company—Coke Battery

Stack:       Scrubber Inlet
Material:
               Acrldlne
Stack Diameter:
                  66"
1975
Date
1/16
1/16













Sampling
Period
Start
13:53
14:24













Stop
14:23
14:54













Test
Mo.
1
2













Sample
Weight
(mg)
<0.660
<0.645













Sampled
Volume
(DSCF)
27.4
27.1













AVERAGE
Stack
Temp.
CP)
40
40













40
Stack Gas
Plowrate
(DSCPM)
99,000
99,000



•









99,000
Peed
Rate*
[tons/hr)
















Concen-
+f& tl on
(ppm)
<0.11
<0.11













<0.11
Emission
Rate
Ib/hr*
<0.32
<0.31













<0.32
Ib/ton
of feed
















  * Assuming continuous  pushing
                                                    Clayton Environmental  Consultants, Inc.

-------
                                         TABLE 6.1.3-5
                                      EMISSIOH TEST SUMMARY
Plant;  Ford Motor 'Company—Coke  Battery

Stack:      Scrubber  Inlet
Material:
Acridine
Stack Diameter:
   66" I.D.
1975
Date
6/11
6/12
6/12












Sampling
Period
Start
13:47
10:37
15:10












A V E ft A G
Stop
14:24
11:15
16:24












Test
No.
1
2
3












E
Sample
Weight
(mg)
<0.018
<0.018
<0.018












Sampled
Volume
(DSCF)
2.17
2.08
2.00













Stack
Temp.
CP)
135
137
183












152
Stack Gas
Plowrate
(DSCPM)
80,300
80,300
80,300












80,300
Peed.
Rate *
(tons/hr)
1097
1079
1053












1076
Concen-
tration
(ppm)
<0.04
<0.04
<0.04












<0.04
Emission
Rate
Ib/hr*
<0.09
<0.09
<0.10












<0.09
Ib/ton
of feed
<0. 00008
<0. 00008
<0. 00009












CO. 00008
                                                                                                      ro
 * Assuming continuous pushing
                                                     Clayton Environmental Consultants, Inc.

-------
                                     TABLE 6.1.3-6
                                  EMISSION TEST SUMMARY
     .. Ford Motor Company--Coke Battery

Stack:
                                             Material:
                                                              Acridine
        Scrubber Outlet
                                             Stack Diameter:
90" I.D.
1975
Date
6/11
6/12
6/12












Sampling
Period
Start
13:47
10:37
15:10












A V E R A G
Stop
14:24
11:15
16:24












Test
No.
1
2
3












E
Sample
Weight
(mg)
<0.018
<0.018
<0.018












Sampled
Volume
(DSCF)
2.71
2.61
2.42













Stack
Temp.
CF)
120
124
122












122
Stack Gas
Flowrate
(DSCFM)
86,900
86,900
86,900












86,900
Feed
Rate*
[tons/hr )
1097
1079
1053












1076
Concen-
(ppm)
<0.03
<0.03
<0.04












<0.03
Emission
Rate
Ib/hr*
<0.08
<0.08
<0.09












<0.08
Ib/toa
of feed
<0. 00007
<0. 00007
<0. 00009












CO. 00008
                                                                                                   NJ
Assuming continuous pushing
                                                 Clayton Environmental  Consultants, Inc.

-------
                                         TABLE 6.1.3-7
                                      EMISSIOH TEST SUMMARY
Plant;   Ford  Motor  Company—Coke  Battery

Stack:   Scrubber  Inlet	
Material:
Acroleln
Stack Diameter:
                  66" I.D.
1975
Date
1/16
1/16













Sampling
Period
Start
13:53
14:24



•









A V I 1 A 6
Stop
14:23
14:54













B
Test
No.
1
2














Sample
Weight
(mg)
<0.10
<0.10













Sampled
Volume
(DSCF)
27.4
27.1














Stack
Temp.
CF)
40
40













40
Stack Gas
Flowrate
(DSCFM)
99,000
99,000













99,000
Feed.
Rate*
(tons/hr)
















Concen-
tration
(ppm)
<0.06
<0.06













<0.06
Emission
Rate
Ib/hr*
<0.05
<0.05













<0.05
Ib/ton
of feed
















                                                                                                       N>
                                                                                                       •P-
 * Assuming continuous pushing
                                                     Clayton Environmental Consultants,  Inc,

-------
                                         TABLE 6.1.3-8
                                      EMISSION TEST SUMMARY
giant; Ford  Motor  Company--Coke Battery

Stack:   Scrubber  Outlet	
Material:
Filterable Acroletn
Total Acroleln  	
Stack Diameter:   90"I.O.
1975
Date
Sampling
Period
Start
Stop
Test
No.
FILTERABLE ACROLEIN
7/17
7/17
08 : 24
14:41
08:51
15:15
1
2
Sample
Weight
(mg)

<0.022
<0.023
Sampled
Volume
(DSCf)

2.81
2.67
AVERAGE

TOTAL
7/17
7/17









ACROLEIN
08 : 24
14:41







08:51
15:15









1
2







AVERAGE


<0.094
<0.085









2.81
2.67








Stack
Temp.
<°F)

120
126
123


120
126







123
Stack Gas
Flowrate
(DSCFM)

86,900
86,900
86,900


86,900
86,900







86,900
Feed
Rate*
(tona/hr)

1044
1034
1039


1044
1034







1039
Concen-
t* f A 1 4 An
(ppm )

<0.12
<0.13
<0.12


<0.51
<0.48







<0.50
Emission
Rate
Ib/hr*

<0.09
<0.10
<0.10


<0.38
<0.37







<0.38
Ib/ton
of feed

<0. 000086
<0. 00 00 96
<0. 000091


<0. 00037
<0. 00035







<0. 00036
                                                                                                       N5
                                                                                                       Ul
    Assuming  continuous pushing
                                                     Clayton Environmental Consultants,  Inc.

-------
                                         TABLE 6.1*3-9
                                      EMISSION TEST SUMMARY
Plant: Ford Motor  Company--Goke Battery

Stack:
                                       Material:   Ammonia
Scrubber Inlet
Stack Diameter:   66" I»D'
1975
Date
2/6
2/6
2/6












Sampling
Period
Start
11:25
12:35
14:00












Stop
12:25
13:35
15:00












Test
No.
1
2
3












Sample
Weight
(mg)
0.043
0.060
<0.039












Sampled
Volume
0>SCF)
53.8
54.1
58.9












AVERAGE
Stack
Temp.
CP)
34
202
37












91
Stack Gas
Flovrate
(DSCFM)
71,900
71,900
71,900












71,900
Feed
Rate*
(tons/hr )
















Concen-
tration
(ppm)
0.04
0.06
<0.03












0.03-
0.04
Emission
Rate
Ib/hr*
0.008
0.01
<0.006












0.006-
0.008
Ib/ton
of feed
















                                                                                                      o>
                                                                                                      I
 * Assuming continuous  pushing
                                                     Clayton Environmental Consultants, Inc.

-------
                                          TABLE 6.1.3-10
                                      EMISSION TEST SUMMARY
Pl*nt:_

Stack:
Ford Motor Company--Coke  Battery

 Scrubber Outlet
Material:
                                                                Ammonia
                                        Stack Diameter:
                    90"  I.D.
1975
Date
6/3
6/3













Sampling
Period
Start
10:40
11:14













Stop
11:05
12:05













Test
No.
1
2













Sample
Weight
(mg)
<0.05
<0.04













Sampled
Volume
(liters)
3.09
3.05













AVERAGE
Stack
Temp.
CF)
















Stack Gas
Flowrate
(DSCFM)
86,900
86,900













86,900
Feed
Rate *
(tons/hr)
1018
1042













1030
Concen-
t TT& fc 4 on
(ppm)
<23
<18













<20
Emission
Rate
Ib/hr*
<5
<4













<4
Ib/ton
of feed
<0.005
<0.004













<0.004
                                                                                                      N)
    Assuming continuous  pushing
                                                     Clayton Environmental Consultants,  Inc.

-------
                                         TABLE 6.1.3-11
                                     EMISSION TEST SUMMARY
       Ford  Motor Company--Coke Battery
Stack:   Scrubber  Inlet
                                                Material:
                                                                Anthracene
                                                Stack Diameter:
                                                                   66" I.D.
1975
Date
1/16
1/16












.
Sampling
Period
Start
13:53
14:24













Stop
14:23
14:54













Test
No.
1
2
•












Sample
Weight
(nig)
<0.581
<0.568













Sampled
Volume
(DSCF)
27.4
27.1













AVERAGE
Stack
Temp.
CF)
40
40













40
Stack Gas
Flowrate
(DSCFM)
99,000
99,000













99,000
Feed
Rate*
(tons/hr)
















Concen-
tration
Tppm;
<0.10
<0.10













<0.10
Emission
Rate
Ib/hr*
<0.28
<0.27



•









<0.28
Ib/ton
of feed
















                                                                                                     oo
                                                                                                     t
* Assuming continuous pushing
                                                    Clayton Environmental Consultants, Inc.

-------
                                         TABLE  6.1.3-12
                                      EMISSION  TEST SUMMARY
Plant: Ford  Motor  Company--Coke Battery

Stack:  Scrubber Inlet
Material:   Anthracene
Stack Diameter:
                    66" I.D.
4975
Date
6/11
6/12
6/12












Sampling
Period
Start
13:47
10:37
15:10












Stop
14:24
11:15
16:24












Test
Ho.
1
2
3












Sample
Weight
(mg)

-------
                                          TABLE 6.1.3-13
                                      EMISSION TEST SUMMARY
Plant:  Ford  Motor  Company--Coke  Battery

Stack:   Scrubber Outlet
Material:
             Anthracene
Stack Diameter:   90" I«D»
1975
Date
6/11
6/12
6/12












Sampling
Period
Start
13:47
10:37
15:10












Stop
14:24
11:15
16:24












Test
No.
1
2
3












Sample
Weight
(mg)
<0.018
<0.018
<0.018












Sampled
Volume
(DSCF)
2.71
2.61
2.42












AVERAGE
Stack
Temp.
CP)
120
124
122












122
Stack Gas
Flovrate
(DSCFM)
86,900
86,900
86,900












86,900
Feed
Rate*
(tons/hr)
1097
1079
1053












1076
Concen-
tration
'. (ppm )
<0.03
<0.03
<0.04












<0.03
Emission
Rate
Ib/hr*
<0.08
<0.08
<0.09












<0.08
Ib/ton
of feed
<0. 00007
<0. 00007
<0. 00009












<0.00008
                                                                                                      U)
                                                                                                      o
 * Assuming continuous pushing
                                                     Clayton Environmental Consultants, Inc.

-------
                                     TABLE 6.1.3-14
                                  EMISSION TEST SUMMARY
Stack:
    Ford Motor Company-- COke Battery

       Scrubber Inlet
                                                Material:
                                                               Benzene
Stack Diameter:
                                                                   66" I.D.
1975
Date
6/5
6/5
6/5
6/5
6/6
6/6









Sampling
Period
Start
12:26
12:50
15:22
16:25
11:27
14:20









Stop
12:27
12:51
15:23
16:26
11:51
14:59









Test
Ho.
1
3
5
7
9
11









Sample
Weight
(»g>
0.003
0.007
<0.001
<0.001

-------
                                          TABLE  6.1.3-15
                                      EMISSION VEST SUMMARY
Plant;  Ford Motor Company--Coke Battery

Stack:  Scrubber Outlet
Material:
Benzene
Stack Diameter:  90" I.D.
1975
Data
6/5
6/5
6/5
6/5
6/6
6/6









Sampling
Period
Start
12:26
12:50
15:22
16:25
11:27
14:20









Stop
12:27
12:51
15:23
16:26
11:51
14:59









Test
Mo.
1
3
5
7
9
11









AVERAGE
Sample
Weight
(mg)
<0.001
<0.001

-------
                                          TABLE  6.1*3-16

                                      EMISSION TEST SUMMARY
Plant: Ford Motor  Company--Cofce^Jattery


Stack:
                                        Material:
Benzene & Homologues
Scrubber Inlet
                                        Stack Diameter:
 66" I.D.
1975
Data
1/10
1/10
Sampling
Period
Start


AVERAGE

2/4
2/6
2/6









15:10
11:50
12:35








Stop















Test
No.
1
2


4
8
11








Sample
Weight
(mg)




<0.033
<0.043
<0.048








Sampled
Volume
(liters)




0.703
0.753
0.774








AVERAGE
Stack
Temp.
<°P>
















Stack Gas
Plovrate
(DSCFM)
98,100
98,100
98,100

71,900
71,900
71,900








71,900
Peed
Rate*
(tons/hr )















-
Concen-
&r A f* 4 ft**
(ppm)
0.82
3.6
2.2

<14.4
<17.6
<19.1








<17.0
Emission
Rate
Ib/hr*
0.98
4.2
2.6

<12.7
<15.4
<16.7








<14.9
Ib/ton
of feed
















                                                                                                      u>
                                                                                                      (jj
  * Assuming continuous pushing
                                                     Clayton Environmental  Consultants, Inc,

-------
                                          TABLE 6.1.3-17
                                      EMISSION TEST SUMMARY
Plant;   Ford Motor  Company -- Coke Battery

Stack:    Scrubber Inlet
Material:
Benzene & Homologues
Stack Diameter:
        66" I.D.
1975
Date
6/5
6/5
6/5
6/5
6/6
6/6









Sampling
Period
Start
12:26
12:50
15:22
16:25
11:27
14:20









Stop
12:27
12:51
15:23
16:26
11:51
14:59









Test
Mo.
1
3
5
7
9
11









Sample
Weight
<*g)
0.005
0.159
0.010
0,003
0.010
0.063









Sampled
Volume
(liters)
1.22
1.21
1.22
1.22
3.60
3.59









AVERAGE
Stack
Temp.
<°F>
















Stack Gas
Flovrate
(DSCFM)
80,300
80,300
80,300
80,300
80,300
80,300









80,300
Feed
Rate*
(tons/hr)
1068
1064
1041
1058
1052
1060









1057
Concen-
tration
(ppm)
1.3
40.4
2.5
0.76
0.85
5.4









8.5
Emission
Rate
Ib/hr*
1.2
39.6
2.5
0.74
0.84
5.3









8.4
Ib/ton
of feed
0.0012
0.0372
0.0024
0.00070
0.00079
0.0050









0.0079
 * Assuming continuous pushing
                                                     Clayton Environmental Consultants, Inc.

-------
                                         TABLE 6.1.3-18
                                      EMISSION TEST SUMMARY
Plant;  Ford  Motor  Company--Coke Battery

Stack:  Scrubber Outlet	
Material:
Benzene  & Homologues
Stack Diameter:
      90" i.D.
1975
Date
6/5
6/5
6/5
6/5
6/6
6/6









Sampling
Period
Start
12:26
12:50
15:22
16:25
11:27
14:20









Stop
12:27
12:51
15:23
16:26
11:51
14:59









Test
No.
1
3
5
7
9
11









Sample
Weight
(mg)
0.025
0.034
0.005
0.007
0.007
0.003









Sampled
Volume
(liters;
1.00
0.99
0.99
1.00
2.98
3.00









AVERAGE
Stack
Temp.
<°P)
















Stack Gas
Flowrate
(DSCFM)
86,900
86,900
86,900
86,900
86,900
86,900









86,900
Feed
Rate*
(tons/hr )
1068
1064
1041
1058
1052
1060









1057
Concen-
f-pA #• Ion
(ppm)
7.7
10.6
1.6
2.2
0.72
0.31









3.9
Emission
Rate
Ib/hr"
8.1
11.2
1.6
2.3
0.77
0.33









4.0
Ib/ton
of feed
0.0076
0.0105
0.0016
0.0022
0.00073
0.00031









0.0038
                                                                                                      CO
                                                                                                      in
    Assuming continuous  pushing
                                                     Clayton Environmental Consultants, Inc.

-------
                                         TABLE 6.1.3-19
                                      EMISSION TEST SUMMARY
Plant;   Ford  Motor Company--Coke  Battery

Stack:     Scrubber Islet
Material:    Benzo(a+e)pyrene
Stack Diameter:
                  66"  I.D.
1975
Data
6/11
6/12
6/12












Sampling
Period
Start
13:47
10:37
15:10












Stop
14:24
11:15
16:24






*





Test
No.
1
2
3












Sample
Weight
(mg)
<0.009
<0.009
<0.009












Sampled
Volume
(DSCF}
2.17
2.08
2.00












AVERAGE
Stack
Temp.
CF)
135
137
183












152
Stack Gas
Flovrate
(DSCFM)
80,300
80,300
80,300












80,300
Feed
Rate*
(tons/hr)
1097
1079
1053












1076
Concen-
tration
(ppm)
<0.01
<0.01
<0.02












<0.01
Emission
Rate
lb/hr*
<0.04
<0.05
<0.05












<0.05
Ib/ton
of feed
<0. 00004
<0. 00004
<0. 00005












CO. 00004
 * Assuming continuous pushing
                                                     Clayton Environmental Consultants,  Inc.

-------
                                       TABLE 6.1.3-20
                                    EMISSION TEST SUMMARY
     ;  Ford Motor Company- -Coke Battery

Stack:
                                              Material:    Benzo(a+e)pyrene
       Scrubber Outlet
                                               Stack Diameter:
90" I.D.
1975
Date
6/11
6/12
6/12












Sampling
Period
Start
13:47
10:37.
15:10












Stop
14:24
11:15
16:24












Test
No.
1
2
3












Sample
Weight
(mg)
<0.009
<0.009
<0.009












Sampled
Volume
(DSCF)
2.71
2.61
2.42












AVERAGE
Stack
Temp.
<°F)
120
124
122












122
Stack Gas
Flowrate
(DSCFM)
86,900
86,900
86,900












86,900
Feed.
Rate*
(tons/hr)
1097
1079
1053












1076
Concen-
+ ra + 4 on
(ppm)
<0.01
<0.01
<0.01












<0.01
Emission
Rate
Ib/hr*
<0.04
<0.04
<0.04












<0.04
Ib/ton
of feed
<0. 00004
<0. 00004
<0. 00004












<0. 00004
                                                                                                    UJ
* Assuming continuous  pushing
                                                   Clayton Environmental Consultants,  Inc.

-------
                                          TABLE 6.1.3-21
                                      EMISSION TEST SUMMARY
Plant:  Ford  Motor  Company--Coke  Battery

Stack:  Scrubber  inlet
Material:
Carbazole
Stack Diameter:
                   66" i.D.
1975
Date
1/16
1/16













Sampling
Period
Start
13:53
14:24













A V E R A 6
Stop
14:23
14:54













Test
No.
1
2













E
Sample
Weight
(mg)
<1.320
<1.290













Sampled
Volume
(DSCF)
27.4
27.1














Stack
Temp.
<°F)
40
40













40
Stack Gas
Flovrate
(DSCFM)
99,000
99,000













99,000
Feed
Rate*
(tons/hr)
















Concen-
tration
(ppm)
<0.24
<0.24













<0.24
Emission
Rate
Ib/hr*
<0,63
<0.62













<0.62
Ib/ton
of feed
















                                                                                                       u>
                                                                                                       00
  * Assuming continuous  pushing
                                                     Clayton Environmental Consultants,  Inc.

-------
                                          TABLE 6.1.3-22
                                      EMISSION TEST SUMMARY
Plant;  Ford  Motor  Company—Coke Battery


Stack:  Scrubber  inlet
Material:
Carbazole
Stack Diameter:   66" I.D.
1975
Date
6/11
6/12
6/12












Sampling
Period
Start
13:47
10:37
15:10












Stop
14:24
11:15
16:24












Test
No.
1
2
3












Sample
Weight
(nig)
<0.018
<0.018
<0.018












Sampled
Volume
(PSCP)
2.17
2.08
2.00












AVERAGE
Stack
Temp.
CP)
135
137
183












152
Stack Gas
Flowrate
(DSCFM)
80,300
80,300
80,300












80,300
Feed
Rate*
(tons/hr)
1097
1079
1053












1076
Concen-
(ppm)
<0.04
<0.04
<0.05












<0.04
Emission
Rate
Ib/hr*
<0.09
<0.09
<0.10












<0.09
Ib/ton
of feed
<0. 00008
<0. 00008
<0. 00009












<0. 00008
                                                                                                      u>
                                                                                                      VO
 * Assuming continuous pushing
                                                     Clayton Environmental Consultants,  Inc,

-------
                                          TABLE 6.1.3-23

                                      EMISSION TEST SUMMARY
Plant:  Ford Motor  Company--Coke  Battery


Stack:
                                     Material:
                                                   Carbazole
Scrubber Outlet
                                     Stack Diameter:
90" I.D.
1975
Data
6/11
6/12
6/12












Sampling
Period
Start
13:47
10:37
15:10












Stop
14:24
11:15
16:24












Test
No.
1
2
3












Sample
Weight
(mg)
<0.018
<0.018
<0.018












Sampled
Volume
(DSCF)
2.71
2.61
2.42












AVERAGE
Stack
Temp.
<°F)
120
124
122












122
Stack Gas
Flowrate
(DSCFM)
86,900
86,900
86,900












86,900
Feed.
Rate*
(tons/hr)
1097
1079
1053












1076
Concen-
tration
(ppm)
<0.03
<0.04
<0.04












<0.04
Emission
Rate
Ib/hr*
<0.08
<0.08
<0.09












<0.08
Ib/ton
of feed
<0. 00007
<0. 00007
<0. 00009












CO. 00008
                                                                                                       .p-
                                                                                                       o
   Assuming continuous pushing
                                                     Clayton Environmental Consultants,  Inc.

-------
                                         TABLE 6.1.3-24
                                     EMISSION TEST  SUMMARY
Stack:
       Ford Motor Company --Coke Battery

          Scrubber Inlet
Material:
              Carbon Disulflde
Stack Diameter:
                          66" t.D.
1975
Data
2/4
2/6
2/6












Sampling
Period
Start
15:25
11:50
12:48












Stop















Test
No.
5
9
12












Sample
Weight
(mg>
<0.06
<0.06
<0.07












Sampled
Volume
(liters)
0.703
0.791
0.653












AVERAGE
Stack
Temp.
(.•»>
















Stack Gas
Flowrate
(DSCFM)
71,900
71,900
71,900












71,900
Feed.
Rate*
(tons/hr)
















Concen-
+ r& f~ 1 an
(ppm)
<27.0
<24.0
<33.9












<28.3
Emission
Rate
Ib/hr*
<23.0
<20.4
<28.9












<24.1
Ib/ton
of feed
















 * Assuming  continuous pushing
                                                    Clayton Environmental  Consultants, Inc.

-------
                                          TABLE 6.1.3-25
                                      EMISSION TEST SUMMARY
Plant: Ford  Motor  Company--Coke Battery


Stack:
                                       Material:
                                                      Carbon Dlsulfide
Scrubber Inlet
                                       Stack Diameter:
66" i.D.
1975
Date
6/5
6/5
6/5
6/6
6/6
6/6









Sampling
Period
Start
12:40
14:52
16:14
10:27
13:38
15:10









Stop
12:41
14:53
16:15
10:50
14:11
15:43









Test
No.
2
4
6
8
10
12









Sample
Weight
(mg)
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05









Sampled
Volume
(liters)
1.23
1.22
1.22
2.38
3.57
3.59









AVERAGE
Stack
Temp.
(°F)
















Stack Gas
Flowrate
(DSCFM)
80,300
80,300
80,300
80,300
80,300
80,300









80,300
Feed
Rate*
(tons/hr)
1060
1060
1058
1073
1001
1046









1050
Concen-
tration
(ppm)
<13
<13
<13
<7
<4
<4









<9
Emission
Rate
Ib/hr*
<12
<12
<12
<6
<4
<4









<8
Ib/ton
of feed
<0.011
<0.011
<0.011
<0.006
<0.004
<0.004









JC0.008
                                                                                                       -O
                                                                                                       ro
 * Assuming continuous pushing
                                                     Clayton Environmental Consultants, Inc.

-------
                                         TABLE 6.1.3-26
                                      EMISSION TEST SUMMARY
Plant; Ford  Motor  Company--Coke Battery

Stack:   Scrubber  Outlet
Material:   Carbon Disulfide
Stack Diameter:   90" I.D.
1975
Date
6/5
6/5
6/5
6/6
6/6
6/6









Sampling
Period
Start
12:40
14:52
16:14
10:27
13:38
15:10









Stop
12:41
14:53
16:15
10:50
14:11
15:43









Test
Mo.
2
4
6
8
10
12









Sample
Weight
(mg)
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05









Sampled
Volume
(liters)
0.99
0.98
1.00
1.99
3.00
2.97









AVERAGE
Stack
Temp.
<°P)
















Stack Gas
Flowrate
(DSCFM)
86,900
86,900
86,900
86,900
86,900
86,900









86,900
Feed
Rate*
(tons/hr)
1060
1060
1058
1073
1001
1046









1050
Concen-
t TPA fr Lon
(ppm)
<16
<16
<16
<8
<5
<5









<11
Emission
Rate
Ib/br*
<16
<17
<16
<8
<5
<5









<11
Ib/too
of feed
<0.015
<0.016
<0.015
<0.007
<0.005
<0.005









<0.010
 * Assuming continuous pushing
                                                     Clayton Environmental Consultants,  Inc.

-------
                                          TABLE 6.1.3-27
                                      EMISSION TEST SUMMARY
Plant;   Ford  Motor  Company--Coke  Battery

Stack:     Scrubber  Inlet
Material:
              Carbon Monoxide
Stack Diameter:
                   66" I.D.
1975
Data
2/4
2/4
2/4
2/4
2/6
2/6
2/6
2/6
2/6






Sampling
Period
Start
14:00
14:45
15:00
15:10
13:10
13:10
13:20
13:20
13:30






Stop















Test
No.
1
2
3
4
6
7
8
9
10






Sample
Weight
(mg)















Sampled
Volume















AVERAGE
Stack
Temp.
<-F)
















Stack Gas
Flowrate
(DSCFM)
71,900
71,900
71,900
71,900
71,900
71,900
71,900
71,900
71,900






71,900
Feed
Rate*
(tons/hr)
















Concen-
tration
(ppm)
20
35
15
20
<10
<10
<10
<10
<10






10-16
Emission
Rate
Ib/hr*
6.3
11.0
4.7
6.3
<3.1
<3.1
<3.1
<3.1
<3.1






3.1-4.9
Ib/ton
of feed
















                                                                                                      •p-
                                                                                                      I
 * Assuming continuous pushing
                                                     Clayton Environmental Consultants, Inc.

-------
                                         TABLE 6.1.3*28
                                      EMISSIOH TEST SUMMARY
Plant;  Ford Motor Company--Coke Battery

Stack:  Scrubber Inlet
Material:   Carbon Monoxide
Stack Diameter:
66" I.D.
1975
Date
6/4
6/4
6/4
6/12
7/16
7/16









Sampling
Period
Start
11:04
13:10.
15:31
16:51
10:06
11:37









Stop
11:05
13:11
15:32
16:52
10:07
11:38









Test
No.
2
5
8
11
13
16









Sample
Weight
(mg>















Sampled
Volume















AVERAGE
Stack
Temp.
<•»->
















Stack Gas
Flovrate
(DSCFM)
80,300
80,300
80,300
80,300
80,300
80,300









80,300
Feed
Rate*
(tons/hr)
1100
1082
1053
1052
1014
1026









1054
Concen-
«•«• A f- 4 An
(PP«tt)
65
<10
<10
<10
98
111









46-51
Emission
Rate
Ib/hr*
23
<3.5
<3.5
<3.5
34
39









16-18
Ib/ton
of feed
0.021
<0.0032
<0.0033
<0.0033
0.034
0.038









0.016-
0.017
                                                                                                      U1

                                                                                                      1
 * Assuming continuous pushing
                                                     Clayton Environmental Consultants, Inc.

-------
                                         TABLE 6.1,3-29
                                      EMISSION TEST SUMMARY
Plant:  Ford  Motor Company—Coke Battery

Stack:  Scrubber  Outlet
Material:
Carbon Monoxide
Stack Diameter:
        90" I.D.
1975
Date
6/4
6/4
6/4
6/12
7/16
7/16









Sampling
Period
Start
11:04
13:10
15:31
16:51
10:06
11:37









Stop
11:05
13:11
15:32
16:52
10:07
11:38









Teat
No.
2
5
8
11
13
16









Sample
Weight
















Sampled
Volume













'

AVERAGE
Stack
Temp.
(°F)
















Stack Gas
Flowrate
(DSCFM)
86,900
86,900
86,900
86,900
86,900
86,900









86,900
Feed
Rate *
(tons/hr)
1100
1082
1053
1052
1014
1026









1054
Concen-
tration
(PP»)
<10
<10
<10
<10
<60
<60









<30
Emission
Rate
Ib/hr*
<3.8
<3.8
<3.8
<3.8
<23
<23









<10
Ib/ton
of feed
<0.0034
<0.0035
<0.0036
<0.0036
<0.022
<0.022









<0.010
   Assuming continuous pushing
                                                     Clayton Environmental Consultants, Inc.

-------
                                         TABLE  6.1.3-30
                                      EMISSIOH  TEST  SUMMARY
Plant:  Ford Motor Company--Coke Battery

         Scrubber Inlet
Material:
Chlorine
Stack:
Stack Diameter:
                      66"  i.D.
1975
Date
2/6
2/6
2/6
2/14
2/14
2/14









Sampling
Period
Start
11:30
12:38
14:03
11:18
12:20
13:24









Stop
12:30
13:38
15:03
12:18
13:20
14:24









Test
No.
1
2
3
1
2
3









Sample
Weight
<»g)
<0.0026
0.0040
0.0040
0.0120
<0.0040
0.0058









Sampled
Volume
(DSCF)
50.7
51.4
55.6
56.2
57.0
56 .5









AVERAGE
Stack
Temp.
CF)
34
40
35
40
42
46









40
Stack Gas
Flowrate
(DSCFM)
71,900
71,900
71,900
69,600
69,600
69,600









70,800
Feed.
Rate *
(tons/hr )
















Concen-
ts PA 1 4 nn
(ppm)
<0.0006
0.0009
0.0009
0.003
<0.0008
0.001









0.001
Emission
Rate
Ib/hr*
tO. OOQ5
OU.0007
0.. 000-7
0.002
CO- 0006
0.0009









O.Q007
0.0909
Ib/ton
of feed
















   Assuming continuous  pushing
                                                    Clayton Environmental Consultants, Inc.

-------
                                          TABLE 6.1.3-31

                                      EMISSION TEST SUMMARY
Plant;  Ford  Motor  Company-Poke  Battery


Stack:   Scrubber Outlet
Material:
Chlorine
Stack Diameter:   90"
1975
Date
Sampling
Period
Start
CHLORINE
6/3
6/4












16:40
09:04












A V E R A G
Stop

17:11
09:27












E
Test
No.

1
2













Sample
Weight
(ag)

<0.003
<0.003












Sampled
Volume
GDSCF)

2.95
2.69












•
Stack
Temp.
CP)

120
117












118
Stack Gas
Flowrate
(DSCFM)

86,900
86,900












86,900
Feed.
Rate*
(tons/hr)

1058
1058












1.58
Concen-
tration
(ppm)

<0.01
<0.01












<0.01
Emission
Rate
Ib/hr*

<0.01
<0.01












<0.01
Ib/ton
of feed

<0. 000001
<0.00000<












<0.000009
                                                                                                       -p-
                                                                                                       -oo
 * Assuming continuous pushing
                                                     Clayton Environmental Consultants,  Inc.

-------
                                          TABLE  6.1.3-32
                                      EMISSION TEST SUMMARY
Plant; Ford Motor Company--Coke Battery


Stack:        Scrubber Inlet	
Material:
Chrysene
Stack Diameter:  66"  I.D.
1975
Date
1/16
1/16













Sampling
Period
Start
13:53
14:24













Stop
14:23
14:54













Test
No.
1
2













Sample
Weight
(mg)















Sampled
Volume
(DSCF)
27.4
27.1













AVERAGE
Stack
Temp.
<°F>
40
40













40
Stack Gas
Flovrate
(DSCFM)
99,000
99,000













99,000
Feed.
Rate *
(tons/hr)
















Concen-
(ppm)
<0.40
<0.39













<0.40
Emission
Rate
Ib/hr*
<1.4
<1.4













<1.4
Ib/ton
of feed
















                                                                                                      -o
                                                                                                      VO
   Assuming continuous pushing
                                                     Clayton Environmental Consultants, Inc.

-------
                                          TABLE 6.1.3-33
                                      EMISSION TEST SUMMARY
Plant;  Ford Motor Company--Coke  Battery

Stack:  Scrubber Inlet
Material:
Chrysene
Stack Diameter:
                    66" I.D.
1975
Date
6/11
6/12
6/12












Sampling
Period
Start
13:47
10:37
15:10












Stop
14:24
11:15
16:24












Test
No.
1
2
3












AVERAGE
Sample
Weight
(mg)
<0.009
<0.009
<0.009












Sampled
Volume
(DSCP)
2.17
2.08
2.00












•
Stack
Temp.
<°P)
135
137
183












152
Stack Gas
Plovrate
(DSCPM)
80,300
80,300
80,300












80,300
Feed
Rate*
(tons/hr)
1097
1079
1053












1076
Concen-
tration
(ppm)
<0.02
<0.02
<0.02












<0.02
Emission
Rate
Ib/hr*
<0.04
<0.05
<0.05












<0.05
Ib/ton
of feed
<0. 00004
<0. 00004
<0. 00005












<0. 00004
 * Assuming continuous pushing
                                                     Clayton Environmental Consultants, Inc.

-------
                                         TABLE 6.1.3-34
                                      EMISSION  TEST SUMMARY
Plant;  Ford Motor Company--Coke Battery

Stack:    Scrubber Outlet
Material:
Chrysene
Stack Diameter:
                     90"  I.D.
1975
Date
6/11
6/12
6/12












Sampling
Period
Start
13:47
10:37
15:10












Stop
14:24
11:15
16:24












Test
No.
1
2
3












Sample
Weight
(mg)
<0.009
<0.009
<0.009












Sampled
Volume
(DSCF)
2.71
2.61
2.42












AVERAGE
Stack
Temp.
CF>
120
124
122












122
Stack Gaa
Flovrate
(DSCFM)
86,900
86,900
86,900












86,900
Feed.
Rate*
(tons/far)
1097
1079
1053












1076
Concen-
(ppm)
<0.01
<0.01
<0.01












<0.01
Emission
Rate
Ib/hr*
<0.04
<0.04
<0.04












<0.04
Ib/ton
of feed
<0. 00004
<0. 00004
<0. 00004












CO. 00004
 * Assuming continuous pushing
                                                    Clayton Environmental Consultants, Inc,

-------
                                          TABLE  6.1.3-35
                                      EMISSIOH TEST SUMMARY
Stack:
 Motor Company--Coke Battery

Scrubber Inlet
                                                 Material:
                                                                  Cyanide
                                       Stack Diameter
                                                                    66" I.D.
1975
Data
4/9
4/9
4/10












Sampling
Period
Start
13:21
15t03
13:51












Stop
14:54
15:06
14:44












Test
No.
1
2
3












Sample
Weight
(mg)
<0.00077
0.0012
<0 .00078












Sampled
Volume
(liters)
31.6
2.6
33.8












AVERAGE
Stack
Temp.
(°P)
















Stack Gas
Flowrate
(DSCFM)
90,200
90,200
89,900












90,000
Feed
Rate*
(tons/hr)
















Concen-
tration
(ppm)
<0.02
0.43
<0.02












0.14-
0.16
Emission
Rate
Ib/hr*
<0.008
0.16
<0.008












0.05-
0.06
Ib/ton
of feed
















                                                                                                      Ui
                                                                                                      to
 * Assuming continuous pushing
                                                     Clayton Environmental Consultants,  Inc.

-------
                                         TABLE 6.1.3-36
                                      EMISSION TEST SUMMARY
Plant; Ford Motor Company-Qoke Battery

Stack:   Scrubber Outlet
Material:
Filterable Cyanide
Total Cyanide
Stack Diameter:
                   90" I.D.
1975
Data





Sampling
Period
Start





Stop





FILTERABLE CYANIDE
6/3
6/4
16:40
14:58
AVERAGE


17:11
15 : 20


TOTAL CYANIDE
6/3
6/4
16:40
14:58
AVERAGE


17:11
15:20


Test
No.






1
5



1
5


Sample
Weight
(mg)






0.029
0.048



0.176
0.177


Sampled
Volume
0>SCF)






2.95
2.88



2.95
2.88


AVERAGE
Stack
Temp.
CP)






120
120
120


120
120
120


Stack Gas
Flowrate
(DSCFM)






86,900
86,900
86,900


86,900
86,900
86,900


Feed.
Rate*
(tons/hr)






1058
1026
1042


1058
1026
1042


Concen-
tration
(ppm)






0.32
0.54
0.43


1.9
2.0
2.0


Emission
Rate
Ib/hr*






0.11
0.19
0.15


0.69
0.71
0.70


Ib/ton
of feed






0.00010
0.00019
0.00014


0.00065
0.00069
0.00067


   Assuming continuous pushing
                                                     Clayton Environmental Consultants, Inc.

-------
                                          TABLE 6.1.3-37
                                      EMISSION TEST SUMMARY
Plant: Ford  Motor  Company—Coke  Battery

Stack:   Scrubber  Inlet
Material: Ethylene & Homologues
Stack Diameter:
                   66" I.D.
1975
Date
1/10
1/10
1/10
Sampling
Period
Start



AVERAGE

2/4
2/4
2/4
2/4
2/6
2/6
2/6
2/6
2/6


14:00
14:45
15:00
15:10
13:10
13 : 10
13:20
13:20
13:30

Stop















Test
No.
1
2
3


1
2
3
4
6
7
8
9
10

Sample
Weight
(ing)















Sampled
Volume















AVERAGE
Stack
Temp.
CD
















Stack Gas
Flovrate
(DSCFM)
98,100
98,100
98,100
98,100

71,900
71,900
71,900
71,900
71,900
71,900
71,900
71,900
71,900

71,900
Feed
Rate*
(tons/hr)
















Concen-
tration
(ppm)
<1.5
<1.5
<1.5
<1.5

<12
<12
<12
<12
<12
<12
<12
<12
<12

<12
Emission
Rate
Ib/hr*
<0.64
<0.64
<0.64
<0.64

<3.8
<3.8
<3.8 •
<3.8
<3.8
<3.8
<3.8
<3.8
<3.8

<3.8
Ib/ton
of feed
















 * Assuming continuous pushing
                                                     Clayton Environmental Consultants, Inc.

-------
                                         TABLE 6.1.3-38
                                      EMISSION  TEST  SUMMARY
Plant: F°rd Motor Company--Coke Battery

Stack: Scrubber inlet
Material:
Ethylene & Homologuea
Stack Diameter:
                   66" I.D.
1975
Date
6/4
6/4
6/4
6/12
7/16
7/16









Sampling
Period
Start
10:49
13:00
15:19
16:36
10: *5
12:03









Stop
10:50
13:01
15:20
16:37
10:36
12:04









Test
NO.
1
4
7
10
14
17









AVERAGE
Sample
Weight
(mg)















Sampled
Volume
















Stack
Temp.
CP>
















Stack Gas
Flowrate
(DSCFM)
80,300
80,300
80,300
80,300
80,300
80,300









80,300
Feed
Rate*
(tons/hr)
1096
1074
952
1052
1029
1066









1045
Concen-
tration
(ppm)
4
15
1
<0.5
<0.5
184









34
Emission
Rate
Ib/hr*
1
5
0.3
<0.2
<0.2
64.5









12
Ib/toa
of feed
0.001
0.005
0.0004
<0.0002
<0.0002
0.0606









0.011
 * Assuming continuous pushing
                                                    Clayton Environmental Consultants, Inc.

-------
                                         TABLE 6.1.3-39
                                      EMISSION TEST SUMMARY
Plant: Ford Motor Company—Coke  Battery

Stack:
                                        Material:
Ethylene & Homologues
Scrubber  Outlet
                                        Stack Diameter:
      90" I.D.
1975
Data
6/4
6/4
6/4
6/12
7/16
7/16









Sampling
Period
Start
10:49
13:00
15:19
16:36
10:35
12:03









Stop
10:50.
13:01
15:20
16:37
10:36
12:04









Test
No.
1
4
7
10
14
17









AVERAGE
Sample
Weight
(ing)















Sampled
Volume
















Stack
Temp.
CP)
















Stack Gas
Plowrate
(DSCPM)
86,900
86,900
86,900
86,900
86,900
86,900









86,900
Feed
Rate *
(tons/hr)
1096
1074
952
1052
1029
1066









1045
Concen-
tration
(ppm)
20
1
2
<0.5
2
0.5









4.3
Emission
Rate
Ib/hr*
7
0.4
0.7
<0.2
0.7
0.2









2
Ib/ton
of feed
0.006
0.0003
0.0007
<0.0002
0.0007
0.0002









0.001
                                                                                                     I/I
 * Assuming continuous pushing
                                                     Clayton Environmental Consultants, Inc.

-------
                                         TABLE 6.1.3-40
                                      EMISSION TEST SUMMARY
Plant:  Pord Motor Company--Coke Battery

Stack:
                                       Material:
                                                     Fluoranthene
Scrubber Inlet
Stack Diameter:   66" I»D»
1975
Date
1/16
1/16













Sampling
Period
Start
13:53
14:24













Stop
14:23
14:54













Teat
No.
1
2













Sample
Weight
(mg)
<0.528
<0.516













Sampled
Volume
(DSCP)
27.4
27.1













AVERAGE
Stack
Temp.
CP)
40
40













40
Stack Gas
Plovrate
(DSCPM)
99,000
99,000




-








99,000
Peed
Rate *
(tons/hr)
















Concen-
tration
(ppm)
<0.08
<0.08













<0,08
Emission
Rate
Ib/hr*
<0.25
<0.25













<0.25
Ib/too
of feed
















                                                                                                      tn
   Assuming  continuous pushing
                                                     Clayton Environmental Consultants, Inc.

-------
                                          TABLE  6.1.3-41
                                      EMISSIOB  TEST SUMMARY
Plant: Ford  Motor  Company--Coke  Battery

Stack:   Scrubber  Inlet
Material:
Fluoranthene,
Stack Diameter:  66"I.D.
1975
Data
6/11
6/12
6/12












Sampling
Period
Start
13:47
10:37
15:10












Stop
14:24
11:15
16:24












Test
No.
1
2
3












Sample
Weight
(mg)
<0.009
<0.009
<0.009












Sampled
Volume
(DSCF)
2.17
2.08
2.00












AVERAGE
Stack
Temp.
CF)
135
137
183












152
Stack Gas
Flovrate
(DSCFM)
80,300
80,300
80,300












80,300
Feed
Rate*
(tons/hr)
1097
1079
1053












1076
Concen-
tration
(ppm)
<0.02
<0.02
<0.02












<0.02
Emission
Rate
Ib/hr*
<0.04
<0.05
<0.05












Ib/ton
of feed
<0. 00004
<0. 00004
<0. 00005












<0.05 ko. 00004
                                                                                                      Ul
                                                                                                      00
 * Assuming continuous pushing
                                                     Clayton Environmental Consultants,  Inc.

-------
                                          TABLE 6..1.3-42
                                      EMISSIOB TEST SUMMARY
Plant:  Ford  Motor  Company—Coke Battery
Stack:
         Scrubber  Outlet
Material:   Fluoranthene
Stack Diameter:
                     90" i.D.
1975
Date
6/11
6/12
6/12












Sampling
Period
Start
13:47
10:37
15:10












Stop
14:24
11:15
16 : 24












Test
No.
1
2
3












Sample
Weight
(mg)
<0.009
<0.009
<0.009












Sampled
Volume
(DSCF)
2.71
2.61
2.42












AVERAGE
Stack
Temp.
CF)
120
124
122












122
Stack Gas
Flowrate
(DSCFM)
86,900
86,900
86,900












86,900
Feed
Rate *
(tons/hr)
1097
1079
1053












1076
Concen-
t*T*A t f on
(ppm)
<0.01
<0.01
<0.02












<0.01
Emission
Rate
Ib/hr*
<0.04
<0.04
<0.04












<0.04
Ib/ton
of feed
<0. 00004
<0. 00004
<0. 00004












CO. 00004
                                                                                                       Ui
                                                                                                       VO
   Assuming  continuous pushing
                                                     Clayton Environmental Consultants,  Inc.

-------
                                       TABLE  6.1.3-43
                                   EMISSION TEST SUMMARY
Plant;  Ford  Motor  Company--Coke Battery

Stack:  Scrubber  Inlet
                                              Material:
Fluorene
                                              Stack Diameter:
  66" I.D,
1975
Data
1/16
1/16













Sampling
Period
Start
13:53
14:24













A V E R A 6
Stop
14:23
14:54













Test
No.
1
2













E
Sample
Weight
(«g>
<1.460
<1.428













Sampled
Volume
(DSCP)
27.4
27.1














Stack
Temp.
(°*>
40
40













40
Stack Gas
Flowrate
(DSCFM)
99,000
99,000













99,000
Feed
Rate *
(tons/hr)
















Concen-
tration
(ppm)
<0.27
<0.27













<0.27
Emission
Rate
Ib/hr*
<0.70
<0.69













<0.70
Ib/ton
of feed
















                                                                                                  o
                                                                                                  i
Assuming continuous pushing
                                                  Clayton Environmental Consultants, Inc,

-------
                                      TABLE 6.1.3-44
                                   EMISSION TEST SUMMARY
      Ford  Motor  Company --Coke Battery
       Scrubber  Inlet
                                             Material:
Fluorene
                                              Stack  Diameter:
    66" I.D.
1975
Date
6/11
6/12
6/12












Sasipling
Period
Start
13:47
10:37
15:10












Stop
14:24
11:15
16:24












Test
No.
1
2
3












Sample
Weight
(mg)
<0.009
<0.009
<0.009












Sampled
Volume
(DSCF)
2.17
2.08
2.00












AVERAGE
Stack
Temp.
<8F)
135
137
183












152
Stack Gas
Flowrate
(DSCFM)
80,300
80,300
80,300












80,300
Feed
Rate*
(tons/hr)
1097
1079
1053












1076
Conceit-
(ppm)
<0.02
<0.02
<0.02












<0.02
Emission
Rate
Ib/hr*
<0.04
<0.05
<0.05












<0.05
Ib/ton
of feed
<0. 00004
<0. 00004
<0. 00005












CO. 00004
Assuming continuous  pushing
                                                  Clayton Environmental Consultants,  Inc.

-------
                                          TABLE  6.1Y3-45
                                      EMISSION TEST SUMMARY
Stack:
Ford Motor Company--Coke Battery

  Scrubber Outlet
                                                 Material:
                                                                Fluorene
Stack Diameter:  90" I'D»
1975
Date
6/11
6/12
6/12












Sampling
Period
Start
13:47
10:37
15:10












Stop
14:24
11:15
16:24












Test
No.
1
2
3












Sample
Weight
(nig)
<0.009
<0.009
<0.009












Sampled
Volume
(DSCF)
2.71
2.61
2.42












AVERAGE
Stack
Temp.
(°F>
120
124
122












122
Stack Gas
Flowrate
(DSCFM)
86,900
86,900
86,900












86,900
Feed.
Rate *
(tons/hr )
1097
1079
1053












1076
Concen-
tration
(ppm)
<0«02
<0.02
<0.02












<0.02
Emission
Rate
Ib/hr*
<0.04
<0.04
<0.04












<0.04
Ib/ton
of feed
<0. 00004
<0. 00004
<0. 00004












<0. 00004
                                                                                                      NJ
  Assuming continuous pushing
                                                     Clayton Environmental Consultants, Inc.

-------
                                         TABLE 6.1.3-46
                                      EMISSION TEST SUMMARY
Plant;  Ford  Motor -Company--Coke Battery

Stack:  Scrubber  inlet	
Material:
              9-Fluorenone
Stack Diameter:
66" I.D.
1975
Date
1/16
1/16













Sampling
Period
Start
13:53
14:24













Stop
14:23
14:54













Test
No.
1
2













Sample
Weight
(mg>
<0.880
<0.860













Sampled
Volume
(DSCF)
27.4
27.1













AVERAGE
Stack
Temp.
<°F>
40
40













40
Stack Gas
Flovrate
(DSCFM)
99,000
99,000













99,000
Feed
Rate*
(tons/hr)
















Concen-
£T*A fc 4 on
(ppm)
<0.15
<0.15













<0.15
Emission
Rate
Ib/hr*
<0.42
<0.42













<0.42
Ib/ton
of feed
















   Assuming continuous pushing
                                                     Clayton Environmental Consultants, Inc.

-------
                                          TABEE  6.1.3-47
                                      EMISSION TEST SUMMARY
Plant;  Ford  Motor Company--Coke Battery

Stack:   Scrubber Inlet
Material:
             9-Pluorenone
Stack Diameter:  66"
1975
Date
6/11
6/12
6/12












Sampling
Period
Start
13:47
10:37
15:10












Stop
14:24
11:15
16:24












Test
No.
1
2
3












Sample
Weight
(mg)
<0.009
<0;009
<0.009












Sampled
Volume
(DSCF)
2.17
2.08
2.00












AVERAGE
Stack
Temp.
(°F)
135
137
183












152
Stack Gas
Flovrate
(DSCFM)
80,300
80,300
80,300












80,300
Feed
Rate *
(tons/hr)
1097
1079
1053












1076
Concen-
tration
(ppm)
<0.02
<0.02
<0.02












<0.02
Emission
Rate
Ib/hr*
<0.04
<0.05
<0.05












<0.05
Ib/ton
of feed
<0. 00004
<0. 00004
<0. 00005












CO. 00004
 * Assuming continuous pushing
                                                     Clayton Environmental Consultants, Inc,

-------
                                         TABLE 6.1.3-48
                                     EMISSIOH TEST  SUMMARY
Stack:
Ford Motor Company--Coke Battery

 Scrubber Outlet
                                                Material:
                                                              9-Fluoreaone
                                         Stack  Diameter:
                                                                     90" I.D.
1975
Date
6/11
6/12
6/12












Sampling
Period
Start
13:47
10:37
15:10












Stop
14:24
11:15
16:24












Test
No.
1
2
3












Sample
Weight
(nig)
<0.009
<0.009
<0.009












Sampled
Volume
(DSCF )
2.71
2.61
2.42












AVERAGE
Stack
Temp.
C»>
120
124
122












122
Stack Gas
Flovrate
(DSCFM)
86,900
86,900
86,900












86,900
Feed.
Rate *
(tons/hr)
1097
1079
1053












1076
Concen-
^vfl fr 4 on
(ppm)
<0.02
<0.02
<0.02












<0.02
Emission
Rate
Ib/hr*
<0.04
<0.04
<0.04












<0.04
Ib/ton
of feed
<0. 00004
<0. 00004
<0. 00004












CO. 00004
  * Assuming continuous pushing
                                                    Clayton Environmental  Consultants, Inc.

-------
                                          TABLE 6.1.3-49
                                      EMISSION TEST SUMMARY
Stack:
         Ford Motor  Company--Coke Battery

          Scrubber Inlet
Material: Total Light Hydrocarbons (as Methane)

                   66" I.D.
Stack Diameter:
1975
Date
1/10
1/10
1/10
Sampling
Period
Start



AVERAGE

2/4
2/4
2/4
2/4
2/6
2/6
2/6
2/6
2/6


14:00
14:45
15:00
15:10
13:10
13:10
13:20
13:20
13:30

Stop















Test
No.
1
2
3


1
2
3
4
6
7
8
9
10

Sample
Weight
(mg)















Sampled
Volume















AVERAGE
Stack
Temp.
(°F)
















Stack Gas
Flowrate
(DSCFM)
98,100
98,100
98,100
98,100

71,900
71,900
71,900
71,900
71,900
71,900
71,900
71,900
71,900

71,900
Feed
Rate*
(tons/hr)
















Concen-
tration
(ppm)
14.7
5.3
1.2
7.1

VOID
34
70
48
48
100
123
31
32

61
Emission
Rate
Ib/hr*
3.6
1.3
0.29
1.7

VOID
6.1
12.6
8.6
8.6
18.0
22.1
5.6
5.8

10.9
Ib/ton
of feed
















  Assuming continuous pushing
                                                     Clayton Environmental Consultants, Inc,

-------
                                         TABLE 6.1.3-50
                                      EMISSION TEST SUMMARY
Plant: Ford Motor Company—Coke Battery

Stack:   Scrubber Inlet
Material:    Total  Light Hydrocarbons  '(as Methane)
Stack Diameter:
                    66"  I.D.
1975
Date
6/4
6/4
6/4
6/12
7/16
7/16









Sampling
Period
Start
10:49
13:00
15:19
16:36
10:35
12:03









Stop
10:50.
13:01
15:20
16 :37
10:36
12:04









Test
No.
1
4
7
10
14
17









AVERAGE
Sample
Weight
(mg)















Sampled
Volume
















Stack
Temp.
<"P)
















Stack Gas
Plowrate
(DSCPM)
80,300
80,300
80,300
80,300
80,300
80,300









80,300
Feed
Rate*
(tons/hr)
1096
1074
952
1052
1029
1066









1045
Concen-
+ r& ££on
(ppm)
40
30
4
3
6
514









100
Emission
Rate
Ib/hr*
8.0
6.0
0.8
0.6
1
103









20
Ib/ton
of feed
0.0073
0.0056
0.0008
0.0006
0.001
0.0968









0.019
 * Assuming continuous pushing
                                                     Clayton Environmental Consultants,  Inc.

-------
                                         TABLE 6,1.3.51
                                      EMISSION TEST SUMMARY
     : Ford Motor Company--Coke Battery

Stack:
                                       Material:   Total Light Hydrocarbons (as Methane)
Scrubber Outlet
                                       Stack Diameter:
90" I.D.
1975
Data
6/4
6/4
6/4
6/12
7/16
7/16









Sampling
Period
Start
10:49
13:00
15:19
16:36
10:35
12:03









Stop
10:50 .
13:01
15:20
16:37
10:36
12:04









Test
No.
1
4
7
10
14
17









Sample
Weight
(mg)















Sampled
Volume















AVERAGE
Stack
Temp.
<°P>
















Stack Gas
Flowrate
(DSCFM)
86,900
86,900
86,900
86,900
86,900
86,900









86,900
Feed
Rate *
(tons/hr)
1096
1074
952
1052
1029
1066









1045
Concen-
tration
(ppm)
61
17
23
3
13
11









21
Emission
Rate
Ib/hr*
13
3.7
5.0
0.6
2.8
2.4









4.6
Ib/ton
of feed
0.012
0.0034
0.0053
0.0006
0.0027
0.0022









0.0044
                                                                                                     oo
                                                                                                      i
 * Assuming continuous  pushing
                                                     Clayton Environmental Consultants, Inc.

-------
                                         TABLE 6.1*3-52
                                      EMISSIOH TEST SUMMARY
Plant:_

Stack:
Ford Motor Company--Coke  Battery

 Scrubber Inlet
Material:
Hydrogen Sulflde
                                        Stack Diameter:
                  66"
1975
Date
4/10
4/10
4/10












Sampling
Period
Start
12:30
12:30
13:51












Stop
13:33
13:33
14:44












Test
No.
1
2
3












Sample
Weight
(mg)
<0.02
<0.02
<0.02












Sampled
Volume
(liters)
33.4
33.6
32.9












AVERAGE
Stack
Temp.
CD
















Stack Gas
Flowrate
(DSCFM)
89,900
89,900
89,900












89,900
Feed
Rate*
(tons/hr)
















Concen-
t fA 1 £nn
(ppm)
<0.42
<0.42
<0.43












<0.42
Emission
Rate
Ib/hr*
<0.20
<0.20
<0.20












<0.20
Ib/ton
of feed
















                                                                                                      vo
   Assuming  continuous pushing
                                                     Clayton Environmental Consultants, Inc.

-------
                                         TABLE 6.1.3-53
                                      EMISSION TEST SUMMARY
Plant: Ford  Motor  Company—Coke  Battery

Stack:
                                        Material:   Hydrogen Sulflde
Scrubber Outlet
Stack Diameter: 90" !•*>•
1975
Date
6/3
6/3













Sampling
Period
Start
10:40
11:14













Stop
11:05
12:05













Test
No.
1
2













Sample
Weight

<0.01
<0.01













Sampled
Volume
(liters)
3.08
3.06













AVERAGE
Stack
Temp.
<°F)
















Stack Gas
Flowrate
(DSCFM)
86,900
86,900













86,900
Feed
Rate *
(tons/hr)
1018
1042













1030
Concen-
tration
(ppm)
<2
<2













<2
Emission
Rate
Ib/hr*
<1
<1













<1
Ib/ton
of feed
<0.001
<0.001













<0.001
                                                                                                      o
                                                                                                      I
   Assuming continuous  pushing
                                                     Clayton Environmental Consultants, Inc.

-------
                                          TABLE 6.1.3-54
                                      EMISSIOH TEST SUMMARY
Plant; Ford Motor Company--Coke Battery

Stack:   Scrubber Inlet
Material:   Methane & Homologues
Stack Diameter:
66" I.D.
1975
Date
1/10
1/10
1/10
Sampling
Period
Start



AVERAGE

2/4
2/4
2/4
2/4
2/6
2/6
2/6
2/6
2/6


14:00
14:45
15:00
15:10
13:10
13:10
13:20
13:20
13:30

Stop















Test
Ho.
1
2
3


1
2
3
4
6
7
8
9
10

Sample
Weight
(nig)















Sampled
Volume















AVERAGE
Stack
Temp.
CF)
















Stack Gas
Flowrate
(DSCFM)
98,100
98,100
98,100
98,100

71,900
71,900
71,900
7?.,900
71,900
71,900
71,900
71,900


71,900
Feed.
Rate *
(tons/hr)
















Concen-
t r& fr 4. on
(ppm)
6.6
5.3
1.5
4.5

15
15
14
19
14
12
17
>.l
10

14
Emission
Rate
Ib/hr*
1.6
1.3
0.37
1.1

2.7
2.7
2.5
3.4
2.5
2.2
3.1
1.6
1.8

2.5
Ib/ton
of feed
















   Assuming  continuous pushing
                                                    Clayton Environmental Consultants, Znc,

-------
                                          TABLE 6.1.3-55
                                      EMISSION TEST SUMMARY
Plant:  Ford  Motor Company—Coke Battery

Stack:   Scrubber  Inlet
Material:
             Methane &Homologues
Stack Diameter:
                     66" I.D.
1975
Date
6/4
6/4
6/4
6/12
7/16
7/16









Sampling
Period
Start
10:49
13:00
15:19
16:36
10:35
12:03









Stop
10:50
13:01
15:20
16:37
10:36
12:04









Teat
No.
1
4
7
10
14
17









Sample
Weight
















Sampled
Volume















A V E 1 A G E
Stack
Temp.
("*)
















Stack Gas
Flovrate
(DSCFM)
80,300
80,300
80,300
80,300
80,300
80,300









80,300
Feed
Rate*
(tons/hr)
1096
1074
952
1052
1029
1066









1045
Concen-
tration
(ppm)
6
13
2
2
3
2









5
Emission
Rate
Ib/hr*
I
3
0.4
0.4
0.6
0.4









1
Ib/ton
of feed
0.001
0.003
0.0004
0.0004
0.0006
0.0004









0.001
  *  Assuming  continuous pushing
                                                     Clayton Environmental Consultants, Inc.

-------
                                         TABLE 6.1.3-56
                                      EMISSION TEST SUMMARY
Plant; Ford Motor Company--Coke Battery

Stack: Scrubber Outlet	
Material:
              Methane & Homologues
Stack Diameter:
                    90" I.D.
1975
Date
6/4
6/4
6/4
6/12
7/16
7/16









Sampling
Period
Start
10:49
13:00
15:19
16:36
10:35
12:03









Stop
10:50
13:01
15:20
16:37
10:36
12:04









Test
No.
1
4
7
10
14
17









Sample
Weight
















Sampled
Volume















AVERAGE
Stack
Temp.
CF>
















Stack Gas
Flowrate
(DSQFM)
86,900
86,900
86,900
86,900
86,900
86,900









86,900
Feed
Rate*
(tons/hr)
1096
1074
952
1052
1029
1066









1045
Concen-
tration
(ppm)
13
6
3
3
2
3









5
Emission
Rate
Ip/hr*
3
1
0.6
0.6
0.4
0.6









1
Ib/ton
of feed
0.003
0.001
0.0006
0.0006
0.0004
0.0006









0.001
                                                                                                      U)
 * Assuming continuous pushing
                                                     Clayton Environmental Consultants, Inc.

-------
                                         TABLE 6.1.3-57
                                      EMISSIOH TEST SUMMARY
Plant S-

Stack:
Ford Motor Company—Coke Battery

   Scrubber Inlet
Material:
                                                                   Naphthalene
                                       Stack Diameter:
                    66" I.D.
1975
Data
1/16
1/16













Sampling
Period
Start
13:53
14:24













Stop
14:23
14:54













Test
No.
1
2













AVERAGE
Sample
Weight
(nig)
<4.4
<4.3













Sampled
Volume
(DSCF.)
27.4
27.1














Stack
Temp.
<°F>
40
40













40
Stack Gas
Flowrate
(DSCFM)
99,000
99,000













99,000
Feed
Rate*
(tons/hr)














-

Concen-
ts i-fl 1 1. on
(ppm)
<1.1
<1.1













<1.1
Emission
Rate
Ib/hr*
<2.1
<2.1













<2.1
Ib/ton
of feed
















   Assuming continuous pushing
                                                     Clayton Environmental Consultants,  Inc.

-------
                                         TABLE 6.1.3-58
                                      EMISSIOH TEST  SUMMARY
Plant; Ford Motor Company--Coke Battery

Stack:   Scrubber Inlet	
Material:
Naphthalene
Stack Diameter:
       66" I.D.
1975
Date
6/11
6/12
6/12












Sampling
Period
Start
13:47
10:37
15:10












Stop
14:24
11:15
16 : 24












Test
No.
1
2
3





-•






Sample
Weight
(mg)
<0.009
<0.009
<0.009












Sampled
Volume
(DSCF)
2.17
2.08
2.00












AVERAGE
Stack
Temp.
CF)
135
137
183












152
Stack Gas
Flovrate
(DSCFM)
80,300
80.300
80.300












80,300
Feed
Rate*
(tons/hr )
1097
1079
1053












1076
Concen-
£T*A ^ion
(ppm)
<0.03
<0.03
<0.03












<0.03
Emissioo
Rate
Ib/hr*
<0.04
<0.05
<0.05











-
<0.05
Ib/ton
of feed
<0. 00004
<0. 00004
<0. 00005












CO. 00004
 * Assuming continuous pushing
                                                    Clayton Environmental Consultants, Inc.

-------
                                          TABLE  6.1.3-59
                                      EMISSION TEST SUMMARY
Plant:  F°rd Motor Company—Coke Battery

Stack:  Scrubber Outlet
Material:  Naphthalene
Stack Diameter:
                   90" I.D.
1975
Data
6/11
6/12
6/12












Sampling
Period
Start
13:47
10:37
15:10












Stop
14:24
11:15
16:24












Test
Ho.
1
2
3












AVERAGE
Sample
Weight
(»g>
<0.009
<0.009
<0.009












Sampled
Volume
(DSCF)
2.71
2.61
2.42













Stack
Temp.
(°F)
120
124
122












122
Stack Gas
Flowrate
(DSCFM)
86,900
86,900
86,900












86,900
Feed
Rate*
(tons/hr)
1097
1079
1053












1076
Concen-
tration
(ppm)
<0.02
<0.02
<0.02












<0.02
Emission
Rate
Ib/hr*
<0.04
<0.04
<0.04












<0.04
Ib/ton
of feed
<0. 00004
<0. 00004
<0. 00004












CO. 00004
 * Assuming continuous pushing
                                                     Clayton Environmental Consultants, Inc.

-------
                                         TABLE 6.1.3-60
                                      EMISSION TEST SUMMARY
Plant: Ford Motor Company--Coke Battery

Stack:   Scrubber Outlet
Material:
Filterable Naphthalene
Total Naphthalene
Stack Diameter:    90" I.D.
1975
Date
Sampling
Period
Start
Stop
Test
No.
FILTERABLE NAPHTHALENE
7/17
7/17
08:24
14:41
08:51
15:15
1
2
Sample
Weight
(mg)

<0.023
<0.024
Sampled
Volume
(DSCF^

2.81
2.67
AVERAGE

TOTAL
7/17
7/17









NAPHTHALENE
08:24
14:41







08:51
15:15









1
2









0.75
<0.089









2.81
2.67







AVERAGE
Stack
Temp.
CF)

120
126
123


120
126







123
Stack Gas
Flowrate
(DSCFM)

86,900
86,900
86,900


86,900
86,900







86,900
Feed.
Rate*
(tons/hr )

1044
1034
1039


1044
1034







1039
Concen-
^r*A £4 on
(ppm)

<0.05
<0.06
<0.06


1.8
<0.22







0.90-
1.0
Emission
Rate
Ib/hr*

<0.09
<0.10
<0.10


3.1
<0.38







1.6-
1.7
Ib/ton
of feed

<0.00009(
<0. 00010
<0. 00010


0.0029
r<0. 00037







0.0014-
0.0016
  * Assuming  continuous pushing
                                                     Clayton Environmental Consultants, Inc.

-------
Plant:,

Stack:
                                          TABLE 6.1.3-61
                                      EMISSIOH TEST SUMMARY
Ford Motor Company--Coke Battery

 Scrubber Inlet
Material:
                                                               Nitrogen Oxides
                                      Stack Diameter:
                   66" I.D.
1975
Data
2/4
2/4
2/4












Sampling
Feriod
Start
13:05
13:23
13:30












Stop
13:06
13:24
13:31












Test
No.
1
2
3












AVERAGE
Sample
Weight
(mg)
0.076
0.155
0.050












Sampled
Volume
(liters)
1.72
1.69
1.87













Stack
Temp.
<°F)
















Stack Gas
Flowrate
(DSCFM)
71,900
71,900
71,900












71,900
Feed
Rate*
(tons/hr)
















Concen-
tration
(ppm)
23.1
47.9
14.0












28.3
Emission
Rate
Ib/hr*
11.9
24.7
7.2












14.6
Ib/ton
of feed
















                                                                                                      00
                                                                                                      1
 *  Assuming  continuous pushing
                                                     Clayton Environmental Consultants,  Inc.

-------
                                         TABLE 6.1.3-62
                                      EMISSIOH TEST SUMMARY
Plant: Ford Motor Company--Coke Battery

Stack:
                                       Material:   Nitrogen Oxides (as
Scrubber Inlet
                                        Stack Diameter:
66" I-D-
1975
Date
6/10
6/10
6/10
6/10
6/10
6/11
6/11
6/11
6/11
7/18
7/18
7/18
7/18


Sampling
Period
Start
14:30
14:58
15:10
16:39
16:57
14:55
15:16
15:28
15:38
12:20
12:27
12:39
12:50


Stop
14:31
14:59
15:11
16:40
16:58
14:56
15:17
15:29
15:39
12:21
12:28
12:40
12:51


Test
No.
1
2
3
4
5
6
7
8
9
10
11
12
13


AVERAGE
Sample
Weight
(mg>
0.04
0.10
0.03
0.02
0.04
0.01
0.08
<0.01
0.01
<0.01
0.01
0.01
0.01


Sampled
Volume
(liters)
1.71
1.68
1.72
1.73
1.43
1.53
1.60
1.60
1.65
1.72
1.73
1.74
1.73



Stack
Temp.
CD
















Stack Gas
Flowrate
(DSCFM)
80,300
80,300
80,300
80,300
80,300
80,300
80,300
80,300
80,300
80,300
80,300
80,300
80,300


80,300
Feed
Rate*
(tons/hr)
1101
1054
1053
1054
1054
1071
1082
1084
1059
1035
1040
1032
1041


1058
Concen-
(ppm)
12.2
31.1
9.1
6.0
14.6
3.4
26.1
<3.3
3.2
<3.0
3.0
3.0
3.0


8.8 -
9.3
Emission
Rate
Ib/hr*
7.0
17.9
5.2
3.5
8.4
2.0
15.0
<1.9
1.8
<1.7
1.7
1.7
1.7


5.1 -
5.3
Ib/ton
of feed
0.0064
0.0170
0.0050
0.0033
0.0080
0.0018
0.0139
<0.0017
0.0017
<0.0017
0.0017
0.0017
0.0017


0.0048-
0.0050
                                                                                                      VO
   Assuming continuous pushing
                                                     Clayton Environmental Consultants,  Inc.

-------
                                        TABLE  6.1.3-63
                                    EMISSION  TEST SUMMARY
Plant:   Ford  Motor  Company--Coke  Battery


Stack:
                                               Material:
Nitrogen Oxides (as NO2)
       Scrubber outlet
                                               Stack Diameter:
       90" I.D.
1975
Date
6/10
6/10
6/10
6/10
6/10
6/11
6/11
6/11
6/11
7/18
7/18
7/18
7/18


Sampling
Period
Start
14:30
14:58
15:10
16:39
16:57
14:55
15:16
15:28
15:38
12:20
12:22
12:39
12:50


Stop
14:31
14:59
15:11
16:40
16:58
14:56
15:17
15:29
15:39
12:21
12:28
12:40
12:51


Test
No.
1
2
3
4
5
6
7
8
9
10
11
12
13


Sample
Weight
Ug)
0.07
0.12
0.03
0.03
0.23
0.03
0.17
0.01
0.01
<0.01
0.01
0.02
0.07


Sampled
Volume
(liters)
1.76
1.81
1.81
1.82
1.80
1.82
1.82
1.84
1.84
1.79
1.65
1.80
1.76


AVERAGE
Stack
Temp.
(°F>
















Stack Gas
Flovrate
(DSCFM)
86,900
86,900
86,900
86,900
86,900
86,900
86,900
86,900
86,900
86,900
86,900
86,900
86,900


86,900
Feed
Rate*
(tons/hr)
no;
1054
1053
1054
1054
1071
1082
1084
1059
1035
1040
1032
1041


1058
Concen-
tration
(ppm)
20.8
34.6
8.7
8.6
66.8
8.6
48.8
2.8
2.8
<2.9
3.2
5.8
20.8


17.9-18.:
Emission
Rate
Ib/hr*
13.0
21.6
5.4
5.4
41.6
5.4
30.4
1.8
1.8 .
<1.8
2.0
3.6
13.0


11.2-
11.3
Ib/ton
of feed
0.0118
0.0205
0.0051
0.0051
0.0395
0.0050
0.0281
0.0016
0.0017
<0.0018
0.0019
0.0035
0.0124


0.0105-
0.0106
                                                                                                     oo
                                                                                                     o
* Assuming continuous pushing
                                                   Clayton Environmental  Consultants, Inc.

-------
                                         TABLE 6.1.3-64
                                      EMISSIOH  TEST SUMMARY
Plant;  Ford  Motor  Company - Coke Battery

Stack:           Scrubber Inlet
Material:
Filterable Partleulate
Condensible Partlculate
Stack Diameter:
       66" I.D.
1975
Date
Sampling
Period
Start
Stop
Test
No.
FILTERABLE P ARTICULATE
4/11
4/14

11:05
13:16

15:30
17:29

1
2

AVERAGE












CONDENSIBLE FARTICULATE
4/11
4/14




11:05
13:16




15:30
17:29




1
2




Sample
Weight
(mg)

636.3
996.8






50.2
25.1




Sampled
Volume
(DSCF)

23.6
24.6






23.6
24.6




AVERAGE
Stack
Temp.
CP)

111
98

104




111
98




104
Stack Gas
Flowrate
(DSCFM)

80,600
90,000

85,300




80.600
90.000




85,300
Feed
Rate*
[tons/hr)
















Concen-
t f A fc £on
(gr/DSCF)

0.416
0.625

0.520




0.033
0.016




0.024
Emission
Rate
Ib/br*

287
482

384




22.7
12.1




17.4
Ib/ton
of feed
















                                                                                                      oo
  *  Assuming continuous pushing
                                                     Clayton Environmental Consultants, Inc.

-------
                                         TABLE 6.1.3-65
                                      EMISSION TEST SUMMARY
Plant: Ford  Motor  Company—Coke Battery

Stack:   Scrubber  Inlet
Material:
              Filterable Fartlculate
Stack Diameter:
                   66" I.D.
1975
Date
6/24
6/25
6/26
6/27
6/30
7/1
7/2
7/2
7/3
7/8
7/10
7/14
7/15
7/16
.
Sampling
Period
Start
09:43
10:05
09:25
10:56
08:54
08:51
09:21
13:27
08:38
09:28
09:40
08:40
08:30
08:24

Stop
14:39
14:29
15:05 '
14:25 '
15:14 .
09:36
09:47
15:46
10:28 .
14:50
15:06
14:25
14:10
13:49

Test
No.
, 1

} 2


> 3

I 4

5
6
7
8
9

AVERAGE
Sample
Weight
(mg)
1883.7

1646.9


2001.3

1040.3

1325.5
1103.3
1347.6
1197.4
1080.4

Sampled
Volume
(DSCF)
16.3

16.3


16.2

11.0

11.3
11.4
11.4
11.1
10.8


Stack
Temp.
C»)
158

183


209

152

137
142
130
143
168

158
Stack Gas
Flovrate
(DSCFM)
82,800

79,700


78,200

79,800

81,700
82,300
81,800
79,400
77,000

80,300
Feed
Rate *
(tons/hr)
1051

1022


1028

1052

1007
1038
981
1006
1035

1024
Concen-
tration
(gr/DSCF)
1.78

1.56


1.91

1.46

1.81
1.49
1.82
1.66
1.54

1.67
Emission
Rate
Ib/hr*
1270

1070


1280

998

1270
1050
1280
1130
1020

1150
Ib/ton
of feed
1.20

1.04


1.24

0.949

1.26
1.01
1.30
1.13
0.984

1.12
                                                                                                      oo
                                                                                                      NJ
 * Assuming continuous pushing
                                                     Clayton Environmental Consultants, Inc.

-------
                                         TABLE 6.1.3-66
                                      EMISSIOH TEST SUMMARY
Plant; Ford Motor  Company—Coke Battery

Stack:   Scrubber  Outlet    	
Material:    Filterable Partlculate
Stack Diameter:   90" I.D.
1975
Date
6/24
6/25
6/26
6/27
6/30
7/1
7/2
7/2
7/3
7/8
7/10
7 /14
7/15
7/16

Sampling
Period
Start
09:43
09:46
09:25
10:40
08:54
08:51
09:21
13:27
08:38
09:28
09:40
08:40
08:30
08:24

Stop
14:27
14:29
15:05
14:25-
15:14-
09:36
09:47
15:46
10:28
14:50
15:06
14:25
14:10
13:49

Test
No.
. 1

> 2


\ »
J
} *
I
5
6
7. :
8
9

Sample
Weight
(mg)
16.4

14.2


15.1

7.9

8.4
9.4
7.6
4.7
6.1

Sampled
Volume
(DSCF)
17.2

17.9


17.9

12.0

12.5
11.9
12.2
11.8
11.4

AVERAGE
Stack
Temp.
CF>
131

127


126

124

131
124
121
124
123

126
Stack Gas
Flowrate
(DSCFM)
86,700

87,800


86,300

88,600

88,800
85,200
88,100
85,700
85,000

86,900
Feed
Rate*
(tons/hr)
1051

1023


1028

1052

1007
1038
981
1006
1035

1025
Concen-
(gr/DSCF)
0.015

0.012


0.013

0.010

0.010
0.012
0.010
0.006
0.008

0.011
Emission
Rate
Ib/hr*
10.9

9.2


9.6

7.7

7.9
8.9
7.3
4.5
6.0

8.0
Ib/ton
of feed
0.0104

0.0090


0.0094

0.0073

0.0078
0.0086
0.0074
0.0045
0.0058

0.0078
                                                                                                     oo
   Assuming  continuous pushing
                                                     Clayton Environmental Consultants, Xnc,

-------
                                          TABLE  6.1.3-67

                                      EMISSION TEST SUMMARY
Plant: Ford Motor Company - Coke Battery


Stack:           Scrubber Inlet
Material:
Condenslble Partlculate
Stack Diameter:
       66" I.D.
1975
Date
6/24
6/25
6/26
6/27
6/30
7/1
7/2
7/2
7/3
7/8
7/10
7/14
7/15
7/16

Sampling
Period
Start
09:43
10:05
09:25
10:56
08:54
08:51
09:21
13:27
08:38
09:28
09:40
08:40
08:30
08:24

Stop
14:39
14:29
15:05
14:25
15:14
09:36
09:47
15:46
10:28
14:50
15:06
14:25
14:10
13:49

Test
No.
\ i
1
1 2
1
)
3

r
)
5
6
7
8
9

Sample
Weight
(tag)
72.3

29.1


69.3

44.8

30.9
41.9
46.6
16.0
10.2

Sampled
Volume
(DSCF)
16.3

16.3


16.2

11.0

11.3
11.4
11.4
11.1
10.8

AVERAGE
Stack
Temp.

-------
                                         TABLE 6.1.3-68
                                     EMISSIOH TEST  SUMMARY
Plant;  Ford Motor Company  -  Coke Battery

Stack:           Scrubber Outlet
Material:
Condenslble Partlculate
Stack Diameter:
       90" I.D.
1975
Date
6/24
6/25
6/26
6/27
6/30
7/1
7/2
7/2
7/3
7/8
7/10
7/14
7/15
7/16

Sampling
Period
Start
09:43
09:46
09:25
10:40
08:54
08:51
09:21
13:27
08:38
09:28
09:40
08:40
08:30
08:24

Stop
14:27
14:29
15:05
14:25
15:14
09:36
09:47
15:46
10:28
14:50
15:06
14:25
14:10
13:49

Test
Mo.
> 1
1
)2
)
)
3
|
I4
)
5
6
7
8
9

Sample
Weight
(mg)
51.4

37.3


39.8

24.1

7.8
10.8
4.4
4.3
10.2

Sampled
Volume
(DSCF)
17.2

17.9


17.9

12.0

12.5
11.9
12.2
11.8
11.4

AVERAGE
Stack
Temp.
CD
131

127


126

124

131
124
121
124
123

126
Stack Gas
Flovrate
(DSCFM)
86.700

87,800


86,300

88,600

88,800
85,200
88,100
85,700
85,000

86,900
Feed.
Rate*
(tons/hr)
1051

1023


1028

1052

1007
1038
981
1006
1035

1025
Concen-
+ m !• ion
(gr/DSCF)
0.046

0.032


0.034

0.031

0.010
0.014
0.006
0.006
0.014

0.021
Emlssioa
Rate
Ib/hr*
34*3

24.2


25.4

23.5

7.3
10.2
4.2
4.1
10.1

15.9
Ib/ton
of feed
0.0326

0.0237


0.0247

0.0224

0.0073
0.0099
0.0043
0.0041
0.0097

0.0154
                                                                                                     oo
    Assuming continuous pushing
                                                    Clayton Environmental Consultants, Inc.

-------
                                         TABLE 6.1.3-69
                                      EMISSION TEST SUMMARY
Plant;  Ford Motor Company - Coke Battery

Stack:   Scrubber Inlet
              Filterable Chloroform-Ether
Material:     Soluble Partlculate.	
Stack Diameter:
66" I.D.
1975
Data
6/24
6/25
6/26
6/27
6/30
7/1
7/2
7/2
7/3
7/8
7/10
7/14
7/15
7/16



Sampling
Period
Start
09:43
10:05
09:25
10:56
08:54
08:51
09:21
13:27
08:38
09:28
09:40
08:40
08:30
08:24



A V B R A G
Stop
14:39
14:29
15:05
14:25
15:14
09:36
09:47
15:46
10:28
14:50
15:06
14:25
14:10
13:49



E
Test
No.
• 1
2
3
4
5
6
7
8
9




Sample
Weight
<«g)
17.6
29.4
42.5
21.8
24.3
8.8
3.1
2.2
4.2




Sampled
Volume'
(DSCF)
16.3
16.3
16.2
11. 0
11.3
11.4
11.4
11.1
10.8




Stack
Temp.
<°F)
158
183
209
152
137
142
130
143
168



158
Stack Gas
Flowrate
(DSCFM)
82,800
79,700
78,200
79,800
81,700
82,300
81,800
79,400
77,000



80,300
Feed
Rate*
(tons/hr)
1051
1022
1028
1052
1007
1038
981
1006
1035



1024
Concen-
tration
(gr/DSCF)
0.017
0.028
0.040
0.031
0.033
0.012
0.004
0.003
0.0061-



0.019
Emission
Rate
Ib/hr*
11.8
19.0
27.1
20.9
23.2
8.4
2.9
2.1
4.0



13.3
Ib/ton
of feed
0.0113
0.0186
0.0264
0.0199
0.0231
0.00 31
0.0030
0.0021
0.0038



0.0129
                                                                                                       oo
  * Assuming continuous pushing
                                                    Clayton Environmental Consultants,  Inc.

-------
                                        TABLE  6.1.3-70
                                     EMISSION TEST SUMMARY
Plant;  Ford Motor Company -  Coke  Battery
Stack:   Scrubber Outlet
            Filterable Chloroform-Ether
Material:   Soluble  Particulate  	
Stack Diameter:
90" I.D.
1975
Dat«
6/24
6/25
6/26
6/27
6/30
7/1
7/2
7/2
7/3
7/8
7/10
7/14
7/15
7/16



Sampling
Period
Start
09:43
09:46
09:25
10:40
08:54
08:51
09:21
13:27
08:38
09:28
09:40
08:40
08:30
08:24



Stop
14:27
14:29
15:05
14:25
15:14
09:36
09:47
15:46
10:28
14:50
15:06
14:25
14:10
13:49



Test
No.
' 1
2
3
4
5
6
7
8
9



Sample
Weight
(mg)
0.9
<0.5
1.3
,0,9
<0.5
<0.5
<0.5
0.7
1.9



Sampled
Volume'
(DSCF)
17.2
17.9
17.9
12.0
12.5
11.9
12.2
11.8
11.4



AVERAGE
Stack
Temp.
<°F>
131
127
126
124
131
124
121
124
123



126
Stack Gas
Flovrate
(DSCFM)
86,700
87,800
86,300
88,600
88,800
85,200
88,100
85,700
85,000



86,900
Feed
Rate*
[tons/hr)
1051
1023
1028
1052
1007
1038
981
1006
1035



1025
Concen-
cracxon
(gr/DSCF)
0.0008
<0.0004
0.001
0.001
<0.0006
<0.0006
<0.0006
0.0009
0.003



0.0007
-0.001
Emission
Rat*
Ib/hr*
0.60
<0.32
0.83
0.88
<0.47
<0.47
<0.48
0.67
1.9



0.54
-0.74
Ib/ton
of feed
0.0006
<0.0003
0.0008
0.0008
<0.0005
<0.0005
<0.0005
0.0007
0.0018



0.0005
-0.0007
                                                                                                       oo
 * Assuming continuous pushing
                                                    Clayton Environmental Consultants,  Inc.

-------
                                         TABLE 6.1.3-71
                                      EMISSIOH TEST SUMMARY
Plant:  Ford  Motor  Company - Coke Battery


Stack: 	Scrubber Inlet	
Material:
Condensible Chloroform -
Ether Soluble Partlculate
Stack Diameter:
               66" I.D.
1975
Date
6/24
6/25
6/26
6/27
6/30
7/1
7/2
7/2
7/3
7/8
7/10
7/14
7/15
7/16

Sampling
Period
Start
09:43
10:05
09 : 25
10:56
08:54
08:51
09:21
13:27
08:38
09:28
09:40
08:40
08:30
08:24

Stop
14:39
14:29
15:05
14:25
15:14
09:36
09:47
15:46
10:28
14:50
15:06
14:25
14:10
13:49

Test
Ho.
} l
}
} 2
}

3

r
)
5
6
7
8
9

Sample
Weight
<«g>
7.4

<0.5


4.8

0.9

2.6
3.3
7.3
4.7
2.8

Sampled
Volume
(DSCF)
16.3

16.3


16.2

11.0

11.3
11.4
11.4
11.1
10.8

AVERAGE
Stack
Temp.
CF)
158

183


209

152

137
142
130
143
168

158
Stack Gas
Flovrate
(DSCFM)
82,800

79,700


78,200

79,800

81,700
82,300
81,800
79,400
77,000

80,300
Feed
Rate*
(tons/hr )
1051

1022


1028

1052

1007
1038
981
1006
1035

1024
Concen-
tration
(gr/DSCP)
0.007

<0.0005


0.005

0.001

0.004
0.004
0.010
0.007
0.004

0.005
Emission
Rate
Ib/hr*
5.0

<0.32


3.1

0.86

2.5
3.2
6.9
4.4
2.6

3.2
Ib/ton
of feed
0.0047

<0.0003


0.0030

0.0008

0.0025
0.0030
0.0071
0.0044
0.0026

0.0031-
0.0032
                                                                                                     00
                                                                                                     oo
 * Assuming continuous poshing
                                                     Clayton Environmental Consultants, Inc.

-------
                                         TABLE 6.1.3-72
                                      EMISSIOH TEST SUMMARY
Plant;  Ford  Motor 'Company - Coke Battery


Stack:	Scrubber Outlet	
Material:
Condensible Chloroform -.
Ether Soluble Particulate
Stack Diameter:
           90" I.D.
1975
Data
6/24
6/25
6/26
6/27
6/30
7/1
7/2
7/2
7/3
7/8
7/10
7/14
7/15
7/16

Sampling
Period
Start
09:43
09:46
09:25
10:40
08:54
08:51
09:21
13:27
08:38
09:28
09:40
08:40
08:30
08:24

Stop
14:27
14:29
15:05
14:25
15:14
09:36
09:47
15:46
10:28
14:50
15:06
14:25
14:10
13:49

Test
No.
1

, 2


• 3

4

5
6
7
8
9

Sample
Weight
(ing)
10.3

1.5


2.7

2.8

1.6
<0.5
0.6
2.7
3.7

Sampled
Volume
(DSCF)
17.2

17.9


17.9

12.0

12.5
11.9
12.2
11.8
11.4

AVERAGE
Stack
Temp.
(•»)
131

127


126

124

131
124
121
124
123

126
Stack Gas
Flovrate
(DSCFM)
86,700

87,800


86,300

88,600

88,800
85,200
88,100
85,700
85,000

86,900
Feed
Rate*
(tons/hr)
1051

1023


1028

1052

1007
1038
981
1006
1035

1025
Concen-
tration
(gr/DSCF)
0.009

0.001


0.002

0.004

0.002
<0.0006
0.0008
0.004
0.005

0.003
Emission
Rate
Ib/hr*
6.9

0.97


1.7

2.7

1.5
<0.47
0.57
2.6
3.6

2.3
Ib/ton
of feed
0.0065

0.0010


0.0017

0.0026

0.0015
<0.0005
0.0006
0.0026
0.0035

0.0022-
0.0023
                                                                                                     oo
                                                                                                     vo
   Assuming continuous pushing
                                                     Clayton Environmental Consultants, Inc.

-------
                                         TABLE 6.1.3-73
                                      EMISSION TEST SUMMARY
plant:   Ford  Motor  Company--Coke Battery

Stack:        Scrubber  inlet
Material:
                 Phenanthrene
Stack Diameter:
                  66" I.D.
1975
Data
1/16
1/16













Sampling
Period
Start
13:53
14:24













Stop
14:23
14:54













Test
No.
1
2













Sample
Weight
(mg>
<0.660
<0.645













Sampled
Volume
(DSCF)
27.4
27.1













AVERAGE
Stack
Temp.
<°F)
40
40













40
Stack Gas
Flowrate
(DSCFM)
99,000
99,000













99,000
Feed
Rate *
(tons/far )
















Concen-
tration
(ppm)
<0.11
<0.11













<0.11
Emission
Rate
Ib/hr*
<0.32
<0.31





-







<0.32
Ib/ton
of feed
















                                                                                                      VO
                                                                                                      o
 * Assuming continuous pushing
                                                     Clayton Environmental Consultants, Inc.

-------
                                        TABLE 6.1.3-74
                                      BHXSSIOH TEST SUMMARY
plant:  Ford  Motor Company--Coke Battery

Stack:
                                     Material:
                                                    Phenanthrene
Scrubber Inlet
                                     Stack Diameter:
66" I.D.
1975
Date
6/11
6/12
6/12












Sampling
Period
Start
13:47
10:37
15:10












Stop
14:24
11:15
16:24












Test
No.
1
2
3












Sample
Weight
(mg)
<0.018
<0.018
<0.018












Sampled
Volume
DSCF)
2.17
2.08
2.00












AVERAGE
Stack
Temp.
CF)
135
137
183












152
Stack Gas
Flovrate
(DSCFM)
80,300
80,300
80,300












80,300
Feed.
Rate*
(tons/hr)
1097
1079
1053












1076
Concen-
(ppm)
<0.04
<0.04
<0.04












<0.04
Emission
Rate
Ib/hr*
<0.09
<0.09
<0.10












<0.09
Ib/ton
of feed
<0. 00008
<0. 00008
<0. 00009












CO. 00008
 * Assuming continuous pushing
                                                     Clayton Environmental Consultants, Inc.

-------
                                         TABLE 6.1.3-75
                                      EMISSION TEST SUMMARY
Plant: Ford  Motor  Company—Coke  Battery

Stack:
                                     Material: Phenanthrene
Scrubber Outlet
                                     Stack Diameter:
90" i.D.
1975
Date
6/11
6/12
6/12












Sampling
Period
Start
13:47
10:37
15:10












A V E E A 6
Stop
14:24
11:15
16:24












Test
No.
1
2
3












E
Sample
Weight
(mg)
<0.018
<0.018
<0.018












Sampled
Volume
(DSCF)
2.71
2.61
2.42













Stack
Temp.
<°F)
120
124
122












122
Stack Gas
Flowrate
(DSCFM)
86,900
86,900
86,900












86,900
Feed
Rate*
(tons/hr)
1097
1079
1053












1076
Concen-
tration
(ppm)
<0.03
<0.03
<0.04












<0.03
Emission
Rate
Ib/hr*
<0.08
<0.08
<0.09












<0.08
Ib/ton
of feed
<0. 00007
<0. 00007
<0. 00009












CO. 00008
                                                                                                     NJ

                                                                                                     I
 * Assuming continuous pushing
                                                     Clayton Environmental Consultants,  Inc.

-------
                                         TABLE  6.1.3.-76
                                      EHXSSZOH  TEST SUMMARY
Stack:
Ford Motor Company--Coke  Battery


   Scrubber Inlet
                                                Material:   Phenol
Stack Diameter:  66" 1
-------
                                         TABLE 6.1.3-77
                                      EMXSSIOH TEST SUMMARY
Plant: Ford  Motor  Company--Coke  Battery

Stack:
                                        Material:
              Filterable Phenolics
              Total Phenolics
Scrubber Inlet
Stack Diameter:  66" I«D»
1975
Data
Sampling
Period
Start
Stop
Test
No.
FILTERABLE PHENOLICS
6/3
6/4
6/4
6/4
6/4
6/5
16:40
09:04
10:49
13:00
14:58
12:26
AVERAGE
17:11
09:27
11:17
13:18
15:20
12:51

TOTAL PHENOLICS
6/3
6/4
6/4
6/4
6/4
6/5
16:40
09:04
10:49
13:00
14:58
12:26
A V B R A G
17:11
09:27
11:17
13:18
15:20
12:51
1
2
3
4
5
6


1
2
3
4
5
6
E
Sample
Weight
(mg)

<0.23
<0.12
<0.13
<0.12
<0.14
<0.15


<0.33
<0.22
<0.23
<0.22
<0.24
<0.25
Sampled
Volume
(DSCF)

2.03
2.10
2.29
2.10
2.17
2.20


2.03
2.10
2.29
2.10
2.17
2.20

Stack
Temp.
CP)

163
135
155
134
138
128
142

163
135
155
134
138
128
142
Stack Gas
Plowrate
(DSCFM)

80,300
80,300
80,300
80,300
80,300
80,300
80,300

80,300
80,300
80,300
80,300
80,300
80,300
80,300
Peed
Rate *
(tons/hr)

1058
1058
1092
1070
1026
1064
1061

1058
1058
1092
1070
1026
1064
1061
Concen-
tration
(ppm)

<1.0
<0.52
<0.51
<0.52
<0.58
<0.61
<0.62

<1.5
<0.94
<0091
<0.94
<1.0
<1.0
<1.0
Emission
Rate
Ib/hr*

<1.2
<0.61
<0.60
<0.61
<0.69
<0.72
<0.74

<1.7
<1.1
<1.1
<1.1
<1.2
<1.2
<1.2
Ib/ton
of feed

<0.0011
<0. 00057
<0. 00055
<0. 00057
<0.00067
<0.00068
<0. 00069

CO. 0016
CO. 0011
CO. 00098
:o.ooio
CO. 0011
C0.0011
:o.oon
                                                                                                      vO
                                                                                                      -P-
  * Assuming  continuous  pushing
                                                     Clayton Environmental Consultants,  Inc,

-------
                                        TABLE 6.1.3-78
                                     EMISSIOH TEST SUMMARY
     .. Ford  Motor  Company--Coke  Battery


Stack:
                                                Material:
                                                           Filterable  Phenolics
                                                           Total  Phenolics
        Scrubber Outlet
                                                Stack Diameter:
90" I.D.
1975
Date
Sampling
Period
Start
Stop
Test
No.
FILTERABLE PHENOLICS
6/3
6/4
6/4
6/4
6/4
6/5
16:40
09:04
10:49
13:00
14:58
12:26
AVERAGE
17:11
09:27
11:17
13:18
15:20
12:51

TOTAL PHENOLICS
6/3
6/4
6/4
6/4
6/4
6/5
16:40
09:04
10:49
13:00
14:58
12:26
17:11
09:27
11:17
13:18
15:20
12:51
1
2
3
4
5
6


1
2
3
4
5
6
AVERAGE
Sample
Weight
(mg)

<0.12
<0.11
<0.12
<0.11
<0.12
<0.ll


<0.22
<0.21
<0.22
<0.21
<0.22
<0.21
Sampled
Volume
(DSCF)

2.95
2.69
2.90
2.83
2.88
2.83


2.95
2.69
2.90
2.83
2.88
2.83

Stack
Temp.
CF>

120
117
118
120
120
122
120

120
117
118
120
120
122
120
Stack Gas
Flovrate
(DSCFM)

86,900
86,900
86,900
86,900
86,900
86,900
86,900

86,900
86,900
86,900
86,900
86,900
86,900
86,900
Feed
Rate*
[tons/hr)

1058
1058
1092
1070
1026
1064
1061

1058
1058
1092
1070
1026
1064
1061
Concen-
tpfl t i on
(ppm)

<0.37
<0.37
<0.37
<0.35
<0.38
<0.35
<0.36

<0.67
<0.70
<0.68
<0.67
<0.69
<0.67
<0.68
Emission
Rate
Ib/hr*

<0.47
<0.47
<0.48
<0.45
<0.48
<0.45
<0.47

<0.86
<0.90
<0.87
<0.85
<0.88
<0.85
<0.87
Ib/ton
of feed

<0. 00044
<0.00044
<0. 00044
<0. 00042
<0. 00047
<0. 00042
<0. 00044

<0. 00081
<0. 00085
<0. 00080
<0. 00080
<0. 00086
<0. 00080
<0. 00082
                                                                                                      VO
                                                                                                      Ol
* Assuming continuous pushing
                                                    Clayton Environmental Consultants, Inc.

-------
                                          TABLE  6.1.3-79
                                      EMISSIOH  TEST SUMMARY
Stack:
Ford Motor -Company--Coke Battery

  Scrubber Inlet
                                                 Material:
                 Pyrene
Stack Diameter:
66" I.D.
1975
Data
1/16
1/16













Sampling
Period
Start
13:53
14:24













A V E R A 6
Stop
14:23
14:54













Test
Mo.
1
2













E
Sample
Weight
(«g>
<0.440
<0.430













Sampled
Volume
(DSCF)
27.4
27.1














Stack
Temp.
(°F)
40
40













40
Stack Gas
Flowrate
(DSCFM)
99,000
99,000













99,000
Feed
Rate*
(tons/hr )
















Concen-
tration
(ppm)
<0.07
<0.07













<0.07
Emission
Rate
Ib/hr*
<0.21
<0.21













<0.21
Ib/ton
of feed
















                                                                                                      VO
                                                                                                      CT>
 * Assuming continuous pushing
                                                     Clayton Environmental Consultants,  Inc.

-------
                                         TABLE 6.1.3-80
                                      EMISSION TEST SUMMARY
Plant: Ford Motor Company--Coke Battery

Stack:
                                      Material:
Pyrene
Scrubber Inlet
                                       Stack Diameter:
             66" i.D.
1975
Date
6/11
6/12
6/12












Sampling
Period
Start
13:47
10:37
15:10












Stop
14:24
11:15
16:24












Test
No.
1
2
3












Sample
Weight
(mg)
<0.009
<0.009
<0.009












Sampled
Volume
(DSCF)
2.17
2.08
2.00












AVERAGE
Stack
Temp.
(°F)
135
137
183












152
Stack Gas
Flowrate
(DSCFM)
80,300
80,300
80,300












80,300
Feed.
Rate*
(tons/hr)
1097
1079
1053












1076
Concen-
t T*A 1 4 on
(ppm)
<0.02
<0.02
<0.02












<0.02
Emission
Rate
Ib/hr*
<0.04
<0.05
<0.05












<0.05
Ib/ton
of feed
<0. 00004
<0. 00004
<0. 00005












CO. 00004
                                                                                                      VO
                                                                                                      •vj
   Assuming continuous pushing
                                                     Clayton Environmental Consultants, Inc.

-------
                                          TABLE 6*1.3-81
                                      EMISSION TEST SUMMARY
Plant: Ford  Motor  Company--Coke  Battery

Stack:     Scrubber  Outlet
Material:
Pyrene
Stack Diameter:    90"I.P.
1975
Date
6/11
6/12
6/12












Sampling
Period
Start
13:47
10:37
15:10












Stop
14:24
11:15
16:24












Test
No.
1
2
3












Sample
Weight
- 
-------
                                         TABLE 6.1.3-82
                                      EMISSION TEST SUMMARY
Plant:  Ford Motor -Company--Coke Battery

Stack:  Scrubber  inlet
Material:
Pyridlne
Stack Diameter:
                   66"  I.D.
1975
Date
2/4
2/6
2/6












Sampling
Period
Start
15:25
11:50
12:48












Stop















Test
No.
5
9
12












Sample
Weight
(mg>
<0.005
<0.005
<0.006












Sampled
Volume
(liters)
0.703
0.791
0.653












AVERAGE
Stack
Temp.
CP)
















Stack Gas
Flovrate
(DSCFM)
71,900
71,900
71,900












71,900
Feed
Rate*
(tons/hr )
















Concen-
£v*A ^ t on
(ppm )
<2.2
<1.9
<2.8






•





<2.3
Emission
Rate
Ib/hr*
<1.9
<1.7
<2.5












<2.0
Ib/ton
of feed
















                                                                                                     \o
                                                                                                     VO
  * Assuming  continuous pushing
                                                     Clayton Environmental Consultants, Inc.

-------
                                         TABLE 6.1.3-83
                                      EMISSIOH TEST SUMMARY
Plant;  Ford Motor Company--Coke Battery

Stack:  Sferubber inlet	
Material:
Pyridine
Stack Diameter:
       66" I.D.
1975
Dete
6/5
6/5
6/5
6/5
6/6
6/6









Sampling
Period
Start
12:26
12:50
15:22
16:25
11:27
14:20









Stop
12:27
12:51
15:23
16:26
11:51
14:59









Test
No.
1
3
5
7
9
11









Sample
Weight
(ag)
<0.001
<0.001
<0.001
<0.001
<0.001
<0.001









Sampled
Volume
(liters)
1.22
1.21
1.22
1.22
3.60
3.59









AVERAGE
Stack
Temp.
CF)
















Stack Gas
Flowrate
(DSCFM)
80,300
80,300
80,300
80,300
80,300
80,300









80,300
Feed
Rate*
(tons/hr)
1068
1064
1041
1058
1052
1060









1057
Concen-
tration
(ppm)
<0.2
<0.2
<0.2
<0.2
<0.08
<0.08









<0.2
Emission
Rate
Ib/hr*
<0.2
<0.2
<0.2
<0.2
<0.08
<0.08









<0.2
Ib/ton
of feed
<0.0002
<0.0002
<0.0002
<0.0002
<0. 00008
<0. 00008









:o.ooo2
                                                                                                      NJ
                                                                                                      O
                                                                                                      O
   Assuming continuous  pushing
                                                    Clayton Environmental  Consultants, Inc.

-------
                                         TABLE  6.1.3-84
                                      EMISSIOH  TEST SUMMARY
Plant; Ford Motor Company—Coke Battery


Stack:     Scrubber Outlet  	
Material:
Pyridtne
Stack Diameter:
                  90"
1975
Data
6/5
6/5
6/5
6/5
6/6
6/6









Sampling
Period
Start
12:26
12:50
15:22
16:25
11:27
14:20









Stop
12:27
12:51
15:23
16:26
11:51
14:59









Teat
No.
1
3
5
7
9
11









Sample
Weight
(mg)
<0.001
<0.001
<0.001
<0.001
<0.001
<0.001









Sampled
Volume
(liters)
1.00
0.99
0.99
1.00
2.98
3.00









AVERAGE
Stack
Temp.
(•P)
















Stack Gas
Flovrate
(DSCFM)
86,900
86,900
86,900
86,900
86,900
86,900









86,900
Feed.
Rate*
(tons/hr)
1068
1064
1041
1058
1052
1060









1057
Concen-
(ppm)
<0.3
<0.3
<0.3
<0.3
<0.1
<0.1









<0.2
Emission
Rate
Ib/hr*
<0.3
<0.3
<0.3
<0.3
<0.1
<0.1









<0.2
Ib/ton
of feed
<0.0003
<0.0003
<0.0003
<0.0003
<0.0001
<0.0001









CO. 0002
                                                                                                      to
                                                                                                      o
    Assuming continuous pushing
                                                     Clayton Environmental Consultants, Inc.

-------
                                      TABLE  6.1.3-85
                                  EMI8SIOH  TEST SUMMARY
Plant:  Ford Motor  Company--Coke Battery

Stack:
                                             Material:
                                                              Sulfur Dioxide
     Scrubber inlet
                                             Stack Diameter:
66" I.D.
1975
Data
2/12
2/12
2/12


'-









Sampling
Period
Start
11:16
12:18
13:28


'









Stop
12:16
13:18
14:28
:
- -










Test
Ho.
1
2
3


.-^ .
-








AVERAGE
Sample
Weight
(mg)
3.72
6.85
10.6
.. " 't ' -
*• f-
- .- » .^ - •


-•






Sampled
Volume
(DSCF)
46.8
46.4
45.9
' *
. .- -
i '•;
. - .









Stack
Temp.
CP)
45
42
44




. -,







44
Stack Gas
Plovrate
(DSCFM)
69,600
69,600
69,600
- .

v









69,600
Peed.
Rate *
(tona/br)


" ;*













Concen-
tration
(ppm)
1.1
2.0
3.1


• - •.
- -• -

..
•





2.1
Emission
Rate
Ib/hr*
0.73
1.4
2.1
•- •- •



, •
-- . .






1.4
Ib/ton
of feed

- '•
- - - - — . --
T
4
-

•--
, - .... .- .
.. -






Assuming continuous pushing
                                                 Clayton Environmental Consultants,  Inc,

-------
                                         TABLE 6.1.3-86
                                      EMISSIOH TEST SUMMARY
plant:  Ford Motor Company--Coke Battery

Stack:  Scrubber  Inlet
Material:
              Sulfur Dioxide
Stack Diameter:
                    66"  I.D.
1975
Data
6/5
6/6
6/6
6/6
6/6
6/9
6/9
6/9
6/10






Sampling
Period
Start
15:02
10:04
11:27
13:38
14:58
10:58
13:35
14:32
09:33






Stop
16:26
10:50
11:51
14:11
15:23
11:29
14:04
15:03
10:15






Test
No.
1
2
3
4
5
6
7
8
9






Sample
Weight

148
139
130
140
160
165
154
179
165






153
Stack Gas
Flowrate
(DSCFM)
80,300
80,300
80,300
80,300
80,300
80,300
80,300
80,300
80,300






80,300
Feed.
Rate*
(tons/hr)
1053
1069
1052
1001
1057
1038
1050
1039
1060






1047
Concen-
(ppm)
11.0
10.0
9.4
11.3
12.5
15.1
12.0
7.5
15.6






11.6
Emission
Rate
Ib/hr*
8.8
8.0
7.5
9.1
10.0
12.1
9.6
6.0
12.5






9.3
Ib/ton
of feed
0.0084
0.0075
0.0072
0.0091
0.0095
0.0117
0.0091
0.0058
0.0118






0.0089
                                                                                                      Is}
                                                                                                      O
 * Assuming continuous pushing
                                                     Clayton Environmental Consultants,  Inc.

-------
                                          TABLE 6,1.3-87

                                      EMISSIOH  TEST SUMMARY
Plant:  Ford  Motor  Company--Coke  Battery


Stack:   Scrubber Outlet
Material:
Sulfur Dioxide
Stack Diameter:
                  90" I.D.
1975
Data
6/5
6/6
6/6
6/6
6/6
6/9
6/9
6/9
6/10





.
Sampling
Period
Start
15:02
10:04
11:27
13:38
14:58
10:58
13:22
14:32
09:33






Stop
16:26
10:50
11:51
14:11
15:23
11:29
14:04
15:03
10:15






Test
No.
1
2
3
4
5
6
7
8
9






Sample
Weight
(«g>
1.47
0.90
0.90
1.12
0.90
1.51
1.15
1.82
1.98






Sampled
Volume
(DSCF)
3.72
2.79
3.04
2.83
2.79
1.78
2.87
2.80
2.91






AVERAGE
Stack
Temp.
<°F)
126
117
120
120
120
126
125
129
128






123
Stack Gas
Flowrate
(DSCFM)
86,900
86,900
86,900
86,900
86,900
86,900
86,900
86,900
86,900






86,900
Feed
Rate *
(tons/far)
1053
1069
1052
1001
1057
1038
1049
1039
1060






1046
Concen-
tration
(ppm)
5.2
4.3
3.9
5.2
4.3
11.2
5.3
8.6
9;o






6.3
Emission
Rate
Ib/hr*
4.5
3.7
3.4
4.6
3.7
9.8
4.6
7.5
7.8






5.5
Ib/ton
of feed
0.0043
0.0035
0.0032
0.0045
0.0035
0.0094
0.0044
0.0072
0.0074






0.0053
                                                                                                      to
                                                                                                      o
                                                                                                      -p-
  * Assuming continuous pushing
                                                     Clayton Environmental Consultants,  Inc.

-------
                                         TABLE 6.1.3-88
                                      EMISSION TEST SUMMARY
Plant:  Ford Motor Company—Coke Battery


Stack:   Scrubber inlet
Material:
Sulfur Trloxlde
Stack Diameter:
                          66"  I.D.
1975
Daea
2/12
2/12
2/12












Sampling
Period
Start
11:16
12:18
13:28












Stop
12:16
13:18
14:28












Teat
No.
1
2
3












Sample
Weight
(tag)
0.681
<0.280
1.20












Sampled
Volume
(DSCP )
46.8
46.4
45.9












AVERAGE
Stack
Temp.
CF)
45
42
44












44
Stack Gas
Flowrate
(DSCFM)
69,600
69,600
69,600












69,600
Feed.
Rate *
(tons/hr)
















Concen-
(ppm)
0.15
<0.06
0.28












0.14-
0.16
Emission
Rate
Ib/hr*
0.13
<0.06
0.24












0.12-
0.14
Ib/ton
of feed
















                                                                                                      l-o
                                                                                                      o
 * Assuming  continuous pushing
                                                     Clayton Environmental Consultants, Inc.

-------
                                         TABLE 6.1.3-89
                                      EMISSION TEST SUMMARY
Plant;  Ford Motor Company—Coke Battery

Stack:  Scrubber  inlet
Material:  Sulfur Trioxide
Stack Diameter:
                 66"
1975
Data
6/5
6/6
6/6
6/6
6/6
6/9
6/9
6/9
6/10





.
Sampling
Period
Start
15:02
10:04
11:27
13:38
14:58
10:58
13:35
14:32
09:33






Stop
16:26
10:50
11:51
14:11
15:23
11:29
14:04
15:03
10:15






Test
No.
1
2
3
4
5
6
7
8
9






Sample
Height
(mg)
<0.16
0.24
<0.12
<0.12
<0.12
<0.48
<0.20
2.40
<0.20






Sampled
Volume
(DSCP)
3.74
2.85
2.79
2.81
2.82
2.73
2.80
2.79
2.86






AVERAGE
Stack
Temp.
CF)
148
139
130
140
160
165
154
179
165






153
Stack Gas
Flowrate
(DSCFM)
80,300
80,300
80,300
80,300
80,300
80,300
80,300
80,300
80,300






80,300
Feed
Rate *
(tons/hr)
1053
1069
1052
1001
1057
1038
1050
1039
1060






1047
Concen-
tration
(ppm)
<0.45
0.89
<0.46
<0.45
<0.45
<1.9
<0.76
9.1
<0.74






1.1-
1.7
Emission
Rate
Ib/hr*
<0.45
0.89
<0.46
<0.45
<0.45
<1.9
<0.76
9.1
<0.74






1.1-
1.7
Ib/ton
of feed
<0. 00043
0.00084
<0. 00043
<0. 00045
<0. 00043
<0.0018
<0. 00072
0.0088
CO. 00070






0.0011
0.0016
    Assuming continuous  pushing
                                                     Clayton Environmental Consultants, Inc.

-------
                                        TABLE 6.1.3-90
                                      EMISSION TEST SUMMARY
Plant:  Ford Motor Company—Coke Battery


Stack:
                                        Material:
                                                      Sulfur Trioxide
Scrubber Outlet
                                        Stack Diameter:
90" I.D.
1975
Date
6/5
6/6
6/6
6/6
6/6
6/9
6/9
6/9
6/10






Sampling
Period
Start
15:02
10:04
11:27
13:38
14:58
10: 58
13:22
14:32
09:33






Stop
16:26
10:50
11:51
14:11
15:23
11:29
14:04
15:03
10:15






Test
No.
1
2
3
4
5
6
7
8
9






Sample
Weight
(mg)
<0.20
<0.16
<0.16
<0.32
0.64
<0.24
<.020
<0.20
<0.24






Sampled
Volume
(DSCF)
3.72
2,79
3.04
2.83
2.79
1.78
2.87
2.80
2.91






AVERAGE
Stack
Temp.
CF)
126
117
120
120
120
126
125
129
128






123
Stack Gas
Flowrate
(DSCFM)
86,900
86,900
86,900
86,900
86,900
86,900
86,900
86,900
86,900






86,900
Feed
Rate *
(tons/hr)
1053
1069
1052
1001
1057
1038
1049
1039
1060






1046
Concen-
tration
(ppm)
<0.57
<0.61
<0.56
<1.2
2.4
<1.4
<0.74
<0.76
<0.88






O.Z7-
1.0
Emission
Rate
Ib/hr*
<0.62
<0.66
<0.61
<1.3
2.6
<1.6
<0.80
<0.82
<0.95






O.Z9-
1.1
lb/ton
of feed
<0. 00055
<0. 00062
<0. 00058
<0.0013
0.0025
<0.0015
<0.0007(
<0.0007<
<0. 00085






O.UUO3-
0.0011
                                                                                                      S3
                                                                                                      o
   Assuming continuous pushing
                                                     Clayton Environmental Consultants,  Inc.

-------
Plant:  Ford  Motor  Company--Coke Battery


Stack:
                                 TABLE 6.1.3-91

                              EMISSION TEST SUMMARY
                                                   Filterable Sulfur Oxides  (as sulfate)

                                        Material:  Total  Sulfur Oxides  (as sulfate)	
Scrubber Inlet
                                        Stack Diameter:
66" I.D.
1975
Data
Sampling
Period
Start
Stop
Teat
No.
Sample
Weight
<»g>
Sampled
Volume
(DSCF)
FILTERABLE SULFUR OXIDES (AS SULFATE)
6/3
6/4
6/4
6/4
6/4
6/5
16:40
09:04
10:49
13:00
14:58
12:26
17:11
09:27
11:17
13:18
15:20
12:51
1
2
3
4
5
6
26.6
10.5
2.9
2.6
4.7
6.8
2.03
2.10
2.29
2.10
2.17
2.20
AVERAGE
TOTAL SULFUR OXIDES (AS SULFATE)
6/3
6/4
6/4
6/4 .
6/4
6/5
16:40
09:04
10:49
13:00
14:58
12:26
17:11
09:27
11:17
13:18
15:20
12:51
1
2
3
4
5
6
A V E 1 A 6 E
46.2
17.9
8.3
8.8
12.3
12.4

2.03
2.10
2.29
2.10
2.17
2.20

Stack
Temp.
<°F)

163
135
155
134
138
128
142

163
135
155
134
138
128
142
Stack Gas
Flowrate
(DSCFM)

80,300
80,300
80,300
80,300
80,300
80,300
80,300

80,300
80,300
80,300
80,300
80,300
80,300
80,300
Feed.
Rate*
(tons/hr)

1058
1058
1092
1070
1026
1064
1061

1058
1058
1092
1070
1026
1064
1061
Concen-
tration
(ppm)

11.6
44.2
11.2
10.9
19.1
27.3
38.1

201.
75.3
32.0
37.0
50.1
49.8
74.2
Emission
Rate
Ib/hr*

139
53.1
13.5
13.2
23.0
32.9
45.8

242
90.6
38.5
44.5
60.2
59.9
89.3
Ib/ton
of feed

0.131
0.0502
0.0123
0.0123
0.0224
0.0309
0.0432

0.229
0.0856
0.0353
0.0416
0.0587
0.0563
0.0844
                                                                                                      to
                                                                                                      o
                                                                                                      oo
 * Assuming continuous pushing
                                                     Clayton Environmental Consultants,  Inc,

-------
                                         TABLE  6.1.3-92
                                      EMISSION TEST SUMMARY
Plant:  Ford Motor Company—Coke Battery

Stack:  Scrubber Outlet
           Filterable Sulfur Oxides  (as Sulfate)
Material:  Total Sulfur Oxides  (as Sulfate)	
Stack Diameter:
90" I.D.
1975
Date
Sampling
Period
Start
Stop
Test
No.
Sample
Weight
(ng>
Sampled
Volume
(DSCF)
FILTERABLE SULFUR OXIDES (as sulfate)
6/3
6/4
6/4
6/4
6/4
6/5
16:40
09:04
10:49
13:00
14:58
12:26
17:11
09:27
11:17
13:18
15:20
12:51
1
2
3
4
5
6
0.76
1.0
1.2
0.46
1.9
1.2
2.95
2.69
2.90
2.83
2.88
2.83
AVERAGE
TOTAL SULFUR OXIDES (as sulfate)
6/3
6/4
6/4
6/4
6/4
6/5
16:40
09:04
10:49
13:00
14:58
12:26
17:11
09:27
11:17
13:18
15:20
12:51
1
2
3
4
5
6
15.96
4.2
3.3
5.46
8.8
6.4
2.95
2.69
2.90
2.83
2.88
2.83
AVERAGE
Stack
Temp.
(•»)

120
117
118
120
120
122
120

120
117
118
120
120
122
120
Stack Gas
Flovrate
(DSCFM)

86,900
86,900
86,900
86,900
86,900
86,900
86,900

86,900
86,900
86,900
86,900
86,900
86,900
86,900
Feed.
Rate*
[tons/hr )

1058
1058
1092
1070
1026
1064
1061

1058
1058
1092
1070
1026
1064
1061
Concen-
fr Y*fl t Ion
(ppm)

2.3
3.3
3.7
1.4
5.8
3.7
3.4

47.8
13.8
10.1
17.0
27.0
20.0
22.6
Emission
Rate
Ib/hr*

3.0
4.3
4.8
1.9
7.6
4.9
4.4

62.2
18.0
13.1
22.2
35.1
26.0
29.4
Ib/ton
of feed

0.0028
0.0040
0.0044
0.0017
0.0074
0.0046
0.0042

0.0588
0.0170
O.Q120
0.0207
0.0343
0.0244
0.0279
                                                                                                      Is)
                                                                                                      O
                                                                                                      VO
   Assuming continuous pushing
                                                     Clayton Environmental Consultants,  Inc.

-------
                                -  210  -
                            TABLE 6.1.4-1
                    SUMMARY OF FUGITIVE EMISSIONS

                         Ford Motor Company
                     Rouge Plant  - Coke  Battery
                         Dearborn, Michigan
1975
Date
6-10
6-10
6-10
6-10
6-10
6-10
6-11
6-11
6-11
6-11
6-11
6-11
6-11
6-12
6-12
6-12
6-12
6-12
6-13
6-13
6-13
6-13
6-13
6-13
6-13
6-13
6-13
6-16
6-16
6-16
6-16
6-16
6-16
6-16
6-16
6-16
6-16
6-16
6-16
6-16
Push
Time
1315
1330
1350
1419
145.8
1510
0905
0938
1000
1423
1445
1455
1516
1410
1418
1434
1510
1521
1115
1129
1325
1341
1404
1438
1452
1512
1536
0930
0940
1015
1027
1039
1050
1101
1110
1123
1130
1330
1352
1402
Oven
No.
42
X52
4
24
44
X54
16
26
36
X61
13
23
43
32
42
X52
14
24
28
38
31
41
X51
13
23
33
43
43
X53
X63
5
15
25
35
45
X55
7
9
19
29
Concentration
(mg/m3)
123
179
24.8
66.6
284
29.4
254
78.4
273 .
35.1
136 *
77.6
632
13.9*
46.0
23.8
37.2
63.7
576
105 *
38.3
290
141
493,
102
143
73.6
47.1
150
243
456
202
221
154
398
379
427
375
195
61.3
Process
Condition
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
1
1
1
Average
Greenness
2.0
2.3
2.0
1.4
3.0
1.2
2.5
1.3
3.0
2.6
2. 3
1.0
3.0
1.8
1.0
1.0
1.0
1.2
3.0
3.0
1.5
2.5
2.5
3.0
2.3
1.5
2.5
1.0
1.0
2.3
3.0
2.8
2.5
1.7
3.0
2.3
2.8
1.3
1.3
1.5
*
   \y U fc. i JU t. 4.  MWX*W.LV1^AAA  WXS  UA V* b &A V W U  f J. ^. M ^ &J. »• W •«—  —- &•> b* »— w hv ••. V ** W • •* •
                                  Clayton Environmental Consultants,Inc

-------
            -  211  -
        TABLE  6.1.4-2
SUMMARY OF FUGITIVE EMISSIONS

     Ford Motor Company
 Rouge  Plant  - Coke Battery
     Dearborn, Michigan
1975
Date
6-16
6-16
6-16
6-16
6-16
6-16
6-17
6-17
6*17
6-17
6-17
6-17
6-17
6-17
6-17
6-17
6-17
6-17
6-17
6-17
6-17
6-17
6-17
6-17
6-18
6-18
6-18
6-18
6-18
6-18
6-18
6-18
6-18
6-18
6-18
6-18
6-18
6-18
6-18
6-18
6-18
Push
Time
1427
1452
1504
1514
1525
1534
0950
1003
1016
1031
1047
1104
1238
1408
1416
1425
1435
1447
1459
1510
1521
1528
1538
1550
0942
0950
1006
1025
1037
1048
1100
1111
1124
1300
1313
1333
1344
1350
1410
1422
1436
Oven
No.
49
X59
2
12
22
32
X62
4
14
24
34
44
16
8
18
28
38
48
X58
1
11
21
31
41
23
33
43
X53
X63
5
15
25
35
27
37
47
X57
9
19
29
39
Concentration
(mg/m3)
673
177
497
44.2
42.5
239
292
516
572
615
196
482
441
472
330
527
600*
339
235
109*
426*
486
585
996*
776
313
355
54.2*
276
886
380
298
429
834
850
184
30.9*
316
93.0
501
256
Process
Condition
1
1
2
2
2
2
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
' 2
2
2
2
2
2
2
Average
Greenness
3.0
1.3
2.0
1.0
1.0
1.3
3.0
3.0
2.3
2.0
2.5
3.0
2.5
2.5
2.5
2.3
2.0
2.0
1.8
2.0
2.3
2.3
2.7
2.5
2.8
2.8
1.8
2.0
2.0
3.0
3.0
2.0
2.8
2.6
3.0
3.0
2.5
2.2
2.5
3.0
1.8

-------
            -  212 -
        TABLE  6.1.4-3
SUMMARY OF FUGITIVE EMISSIONS

     Ford Motor Company
 Rouge  Plant  - Coke Battery
     Dearborn, Michigan
1975
Date
6-18
6-18
6-18
6-19
6-19
6-19
6-19
6-19
6-19
6-19
6-19
6-19
6-19
6-19
6-19
6-19
6-19
6-19
6-19
6-19
6-19
6-20
6-20
6-20
6-20
6-20
6-20
6-20
6-20
6-20
6-20
6-20
6-20
6-20
6-20
6-20
6-20
6-20
6-20
6-20
6-20
Push
Time
1447
1507
1518
1000
1014
1025
1039
1050
1100
1107
1127
1317
1330
1345
1359
1406
1420
1444
1500
1512
1522
0926
0935
0953
1005
1017
1032
1045
1107
1115
1125
1137
1148
1345
1358
1413
1434
1449
1500
1510
1520
Oven
No.
49
X59
2
42
X52
X62
4
14
24
34
44
46
X56
8
18
28
38
48
X58
1
11
X51
X61
3
13
23
33
43
X53
X63
5
15
25
17
27
37
47
X57
9
19
29
Concentration
(mg/m^)
184*
40.7*
269*
210
237*
192
281
595
140
504
569
264
331
496
358
429
341
457
236
121*
443
75.4*
62.5*
556
269
561
589
160
154
345
768
544
219
476
297
385
613
294
636
345
485
Process
Condition
2
2
2
2
2
2
2
2
2
2
,2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
1
1
1
1
1
1
1
1
Average
Greenness
2.6
3.0
2.5
2.0
2.3
1.7
2.3
2.5
1.5
3.0
3.0
2.5
2.5
2.7
3.0
2.7
2.7
2.8
2.7
2.0
3.0
2.2
2.5
3.0
2.0
2.6
3.0
1.4
2.0
2.3
3.0
2.8
1.8
3.0
3.0
2.2
2.4
1.8
1.6
2.0
3.0
              Clayton Environmental  Consultants,Inc

-------
            -  213  -
        TABLE  6.1.4-4
SUMMARY OF FUGITIVE EMISSIONS

     Ford Motor Company
 Rouge  Plant  - Coke Battery
     Dearborn, Michigan
1975
Date
6-23
6-23
6-23
6-23
6-23
6-23
6-23
6-23
6-23
6-23
6-23
6-23
6-23
6-23
6-24
6-24
6-24
6-24
6-24
6-24
6-24
6-24
6-24 ,
6-24
6-24
6-24
6-25
6-25
6-25
6-25
6-25
6-25
6-25
6-25
6-25
6-25
6-25
6-25
6-25
6-25
6-25
6-25
Push
Time
0905
0934
0956
1012
1049
1057
1110
1128
1152
1200
1313
1330
1345
1358
0921
0943
1013
1025
1036
1052
1103
1117
1132
1330
1340
1348
0912
0946
0955
1005
1018
1029
1040
1053
1105
1120
1133
1255
1310
1332
1342
1353
Oven
No.
39
49
X59
2
22
32
42
X52
X&2
4
44
X54
X64
6
X58
1
11
21
31
41
X51
X61
3
X53
X63
5
9
19
29
39
49
X59
-2
12
22
32
42
14
24
34
44
X54
Concentration
(mg/m^)
296
164
105
330*
369
675
577
327
96.8
369
54.3
104
200
268
707
143 *
221
168
487
63.9
44.1
90.5 *
99.9
33.2
197
345
547
221
500
150
152
63.2
84.5*
59.1
501*
453 *
188
390 *
428
146*
73.9
33.2
Process
Condition
2
2
2
2
1
1
1
1
1
1
1
1
1
1
1
2
2
2
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
Average
Greenness
2.3
2.3
1.5
2.7
1.5
2.0
2.5
1.3
1.3
3.0
2.3
1.3
2.0
1.5
2.7
2.5
2.0
1.6
3.0
1.2
1.0
1.0
2.4
1.3
2.2
2.6
3.0
3.0
3.0
1.8
2.3
1.0
3.0
2.0
3.0
3.0
1.8
2.3
2.6
2.2
1.6
1.2
              Clayton Environmental Consultants,Inc

-------
                              -  214 -
                          TABLE  6.1.4-5
                  SUMMARY OF FUGITIVE EMISSIONS

                       Ford Motor Company
                   Rouge  Plant  - Coke Battery
                       Dearborn, Michigan
1975
Date
6-25
6-25
6-25
6-25
6-25
6-26
6-26
6-26
6-26
6-26
6-26
6-26
6-26
6-26
6-26
6-26
6-26
6-26
6-26
6-26
6-26
6-26
6-26
Push
Time
1403
1417
1428
1438
1454
0846
0900
0915
0925
0937
1016
1022
1039
1049
1102
1125
1345
1355
1408
1428
1438
1458
1504
Oven
No.
X64
6
16
26
36
18
28
38
48
X58
1
11
21
31
41
X51
X63
5
15
25
35
45
X55
Concentration
(mg/m3)
154
78.5
548
54.2
126
563
826
138
356
32.6
181*
224*
129
283
58.0*
37.3
59.4
320
129*
123
173*
277*
16.7*
Process
Condition
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
2
2
2
2
3
3
Average
Greenness
2.5
2.3
3.0
2.0
1.4
3.0
3.0
1.8
3.0
1.0
2.8
2.6
2.0
2.6
1.2
1.0
1.0
3.0
2.4
1.8
2.8
2.8
1.6
* Outlier according to methods presented  in Section  6.3.
                                Clayton Environmental Consultants,Inc

-------
           FIGURE 6.2.1

SCHEMATIC DIAGRAM OF MASS BALANCES

        Ford Motor Company
    Rouge Plant - Cok«C Battery
        Dearborn, Michigan
                                                                          ro
                                                                          M
                                                                          in

-------
    -  216  -
1000


ft ft ft
Sftft












•





.
x^





.^




Measured Dust Concentra-
tion Above Hood
(mg/m3)
200
100
90
80
70
80
50
60
30
20
10












-










^







X


^






x


1.0


X







x







X

^xt'


Confidence Limit





x£






/'







^





X
'





Exclv






^
''






id ing
^





•^










Outl:
^

i

. ^
X
^
Con





FIGURE 6.3-1
FUGITIVE DUST CONCENTRATION
MEASURED WITHOUT FUME HOOD
Ford Motor ^Company
Rouge Plant - Coke Battery
Dearborn, Michigan
1 !



1.5 2.0 2.5






.era
"^Inc



'
fiden











•
ludin




ce Li












g Out




nit












liers











3.0
Average Greenness

-------
                    - 217  -
100O
A AA
SPUD'
BOO
700
600
500
400
300
200







/






s
s
/
/


Measured Dust Concentra-
tion Above Hood
(mg/m3)
100
90
80
70
60
50
40
1 30
20
10


/
/




/

//'
//
/




/
/
/




t
/
/
r
/
Exclu<
^
//




/
/
/
J
/

>
/ (
/
/


ling Outlit
s
S
/


/
/
4
/



s
Confidence



/,
\s/\
/ inclu
/
/
s
s
, Confide







rirn^t


s
y
ding Outli

/
/

nee Limit





FIGURE 6-3-2
FUGITIVE DUST CONCENTRATION
MEASURED WITH
FUME HOOD AND FAN OFF
Ford Motor Company
Rouge Plant - Coke Battery
Dearborn, Michigan j






era











1.0
1.5       2.0        2.5
    Average Greenness
3.0

-------
  - 218  -
1UVU


600











Measured Dust Concentra-
tion Above Hood
(mg/m3)
300
I
200
• 1 00
.3 iuu
90
80
70
60
50
40
30
20
10

/
Excl






/


/
/
f
/
/
/
tiding Outl:



/
/
/

^
/
/
/






/
/
/
/
«. //
//
//
//



/
/
/
f

1.0 1.5 2



y
f
/
/
/

/
/




/
/
/
/


/
r
/ Con
f
/



/,
//.

/
/
/ Conf
/
/




Eidence Limit


/
//
/
Including Outliers
/
f


idence Limit




FIGURE 6.3-3
FUGITIVE DUST CONCENTRATION
MEASURED WITH
FUME HOOD AND FAN ON
Ford Motor Company
Rouge Plant - Coke Battery
Dearborn, Michigan


.0 2.5 3.0
Average Greenness

-------
                              -  219  -
                          FIGURE  6.3-4

                     HOOD CAPTURE EFFICIENCY*
               AS A FUNCTION  OF AVERAGE  GREENNESS

                       Ford Motor Company
                   Rouge Plant  -  Coke Battery
                       Dearborn, Michigan
                              Method 2:
                       Engineer-Observer's Estimate
  100


   90


   80


   70


   60


   50


   40


   30


   20


   10




\
Me
of Hoc
(Perce
\
\



thod 1:
d Capture
nt)
V
\
N

\
•v^^
Hood Capture Efficiency ^-\
(Percent)
Based on Revised Fugitive
Emissions Regression









Efficiency


\
\^

^^




0.5 i.o 1.5 2




V
\

^\^









\
\
^\


.0 2.5 3








\
--^


.0
                             Average  Greenness

* The data base used in obtaining these correlations  relies  upon the
  accuracy of novel and unproven techniques  for  fugitive  emission
  estimates;  as such, the correlations must  be used cautiously.

-------
                                                TABLE  7.0-1
                   SUMMARY OF  FILTERABLE  PARTICULATE EMISSIONS AND CONTROL EFFICIENCIES
                                           Ford Motor Company
                                       Rouge Plant - Coke Battery
                                           Dearborn,  Michigan
Test
No.
1
2
3
4

5
6
7
8
9
Date
(1975)
6/24-25
6/26-27
6/30-7/2
7/2-3
*
7/8
7/10
7/14
7/15
7/16
AVERAGE
Average
Green-
ness
. 2.4
2.1
2.3
2.0

1.5
1.8
1.7
1.4
1.4
1.8
Measured
Scrubber
Filterable Particulate
Emissions
Inlet
(Ibs/.
ton- of
feed)
1.20
1.04
1.24
0.949

1.26
1.01
1.30
1.13
0.984
1.12
Outlet
(Ibs/
ton of
feed)
0.0104
0.0090
0.0094
0.0073

0.0078
0.0086
0.0074
0.0045
0.0058
0.0078
Control
Effi-
ciency
99.1
99.1
99.2
99.2

99.4
99.1
99.4
99.6
99.4
99.3
Estimated
Average
Hood Capture
Efficiency (%)
Method
1*
32
37
33
38

47
41
43
48
48
41
' Method •
2**
49
58
52
61

78
67
71
80
80
66
Estimated
Fugitive
. Emissions
(Ibs/ton
of feed)
Based
on
Method
1
2.55
1.77
2.52
1.55

1.42
1.45
1.72
1.22
1.07
1.70
Based
on
Method
2
1.25
0.753
1.14
0.607

0.355
0.497
0.531
0.282
0.246
0.629
Estimated
Emissions
(Ibs/ton
of feed)
Based
on
Method
1
3.75
2.81
3.76
2.50

2.68
2.46
3.02
2.35
2.05
2.82
Based
on
Method
2
2.45
1.79
2.38
1.56

1.62
1.51
1.83
1.41
1.23
1.75
Estimated
Total System
Efficiency
(Percent)
Based
on-
Method
1
32
37
33
38

47
41
43
48
48
41
Based
.en
Method
2
49
57
52
61

78
67
71
80
80
66


ro
NJ
0
1





**Method 2:  Hood capture efficiency estimated froa linear regression on  engineering observer's
              estimate of hood capture efficiency.

-------
                             TABLE 7.0-2
SUMMARY OF CONDENSIBLE PARTICULATE EMISSIONS AND CONTROL EFFICIENCIES

                         Ford Motor Company
                     Rouge Plant - Coke Battery
                         Dearborn, Michigan
Test
No.




1

2

3

4

5
6
7
8
9

Date




6/24-
6/25
6/26-
6/27
6/30-
7/2
7/2-
7/3
7/8
7/10
7/14
7/15
7/16
Average
Average
Green-
ness



2.4

2.1

2.3

2.0

1.5
1.8
1.7
1.4
1-4
1.8
Measured
Scrubber
Condensible
Particulate
Emissions


Inlet
(Ibs/
ton of
feed)
0,0462

0.0184

0.0430

0.0403

0.0293
0.0385
0.0451
0.0150
0.0093
0.0317

Outlet
(Ibs/
ton of
feed)
0.0326

0.0237

0.0247

0.0224

0.0073
0.0099
0.0043
0.0041
0.0097
0.0154

Control
Effi-
ciency
tt)
29.4

0

42.6

44.4

75.1
74.3
90.5
72.7
0
47.7
Estimated
Average
Hood Capture
Efficiency
(Percent)



Method
1*

32

37

33

38

47
41
43
48
48
41



Method
2**

49

58

52

61

78
67
71
80
80
.66
Estimated
Fugitive
Emissions
(Ibs/ton of feed)


Based
on
Method
1
0.0982

0.0313

0.0873

0.0658

0.0330
0.0554
0.0598
0.0162
0.0101
0.0508

Based
on
Method
2
0.0481

0.0133

0.0397

0.0258

0.0083
0.0190
0.0184
0.0038
0.0023
0.0199
Estimated
Push
Emissions
(Ibs/ton of feed)


Based
. on
Method
1
0.1444

0.0497

0.1303

0.1061

0.0623
0.0939
0.1049
0.0312
0.0194
0.0825

Based
on
Method
2
0.0943

0.0317

0.0827

0.0661

0.0376
0.0575
0.0635
0.0188
0.0116
0.0515
Estimated
Total System
Efficiency
(Percent)


Based
on
Method
I
9

0

14

17

35
30
39
35
0
20

Based
on
Method
2
14

0

22

27

59
50
64
58
0
33








N>
N>
I-1
1









* Method 1:. Hood capture efficiency estimated from linear regression on fugitive emissions concentrations.
** Method 2: Hood Capture efficiency estimated from linear regression on engineering' observer's estimate
of hood. capture efficiency. .

-------
                                                TABLE 7.0-3
                   SUMMARY  OF  FILTERABLE  CHLOROFORM-ETHER  SOLUBLE  PARTICULATE  EMISSIONS
                                         AMD CONTROL EFFICIENCIES
                                            Ford Motor  Company
                                       Rouge Plant  - Coke Battery
                                            Dearborn, Michigan
Test
NO.
1
2
3
4
5
6
7
8
9
Date
(1975)
6/24-25
6/26-27
6/30-7/2
7/2-3
7/8
7/10
7/14
7/15
7/16
AVERAGE
Average
Green-
ness
2.4
2.1
2.3
2.0
1.5
1.8
1.7
1.4
1.4
1.8
Measured
Scrubber
Filterable Chloroform-
Ether Soluble
Parttculate Emissions
Inlet
(Iba/..
ton of
feed)
0.0113
0.0186
0.0264
0.0199
0.0231
0.0081
0.0030
0.0021
0.0038
0.0129
Outlet
(Ibs/
ton of
feed)
.0.0006
<0.0003
0.0008
0.0008
<0.0005
<0.0005
<0.0005
t).0007
0.0018
0.0005-
0.0007
Control
Effi-
ciency
ft)
94.7
>98.4
97.0
96.0
>97.8
>93.8
>83.3
66.7
52.6
86.7-89.7
Estimated
Average
Hood Capture
Efficiency
(Percent)
Method
. 1*
32
37
33
38
47
41
43
48
48
41
Method •
2**
49
58
52
61
78
67
71
80
80
66
. Estimated
Fugitive
Emissions
(Ibs /ton
of feed)
Baaed
on
Method
1
0.0240
0.0317
0.0536
0.0325
0.0260
0.0117
0.0040
0.0023
0.0041
0.0211
Based
on
Method
2
0.0118
0.0135
0.0244
0.0127
0.0065
0.0040
0.0012
0.0005
0.0010
0.0084
Estimated
Push
Emissions
(Ibs/ton
of feed)
Based
on
Method
1
0.0353
0.0503
0.0800
0.0524
0.0491
0.0198
0.0070
0.0044
0.0079
0.0340
Baeed
on
Method
0.0231
0.0321
0.0508
0.0326
0.0296
0.0121
0.0042
0.0026
0.0048
0.0213
Estimated
Total System
Efficiency
(Percent)
Based
on-
Method
1
30
36-37
32
36
46-47
38-41
36-43
32
25
35-36
Based
.en
Method
2
46
57-58
50
59
76-78
63-67
60-71
54
42
56-58
                                                                                                             to
                                                                                                             ro
                                                                                                             ro
**Method 2;  Hood capture efficiency  estimated  from  linear  regression on engineering observer's
              estimate of hood  capture  efficiency.

-------
                                               TABLE 7*0-4
      SUMMARY OF CONDENSIBLE CHLOROFORM-ETHER SOLUBLE PARTICIPATE EMISSIONS AND CONTROL EFFICIENCIES
                                           Ford Motor Company
                                       Rouge Plant -  Coke Battery
                                           Dearborn,  Michigan

Test
No.




1

2

3

4

5
6
7
8
9


Date




6/24-
6/25
6/26-
6/27
6/30-
7/2
7/2-
7/3
7/8
7/10
7/14
7/15
7/16
Average

Average
Green-
ness



2.4

2.1

2.3

2.0

1.5
1.8
1.7
1.4
1.4
1.8
Measured
Scrubber Condensable
Chloroform-Ether
Soluble Particulate
Emissions


Inlet
(Ibs/
' ton of
feed)
0.0047

<0.0003

0.0030

0.0008

0.0025
0.0030
0.0071
0.0044
0.0026
0.0031r
0.0032


Outlet
(Ibs/
ton of
feed)
0.0065

0.0010

0.0017

0.0026

0.0015
CO. 0005
0.0006
0.0026
0.0035
0.0022-
0.0023

Control
Effi-
ciency
(*>
0

0

43.3

0

40.0
>83.3
91.5
40.9
0
33-35
Estimated
Average
Hood Capture
Efficiency
(Percent)



Method
1*

32

37

33

38

47
41
43
48
48
41



Method
2**

49

58

52

61

78
67
71
80
80
66
Estimated
Fugitive
Emissions
(Ibs/ton
of feed)


Based
on
Method
1
0.0100

<0.0005

0.0061

0.0013

0.0028
0.0043
0.0094
0.0048
0.0028
0.0046-
0.0047

Based
on
Method
2
0.0049

<0.0002

0.0028

0.0005

0.0007
0.0015
0.0029
0.0011
0.0006
0.0017
Estimated
Push
Emissions
(Ibs/ton
of feed)


Based
on
Method
1
0.0147

<0.0008

0.0091

0.0021

0.0053
0.0073
0.0165
0.0092
0.0054
0.0077-
0.0078

Based
on
Method
2
0.0096

<0.0005

0.0058

0.0013

0.0032
0.0045
0.0100
0.0055
0.0032
0.0048
Estimated
Total
Efficiency
(Percent)


Based
on
Method
1
0

0

14

0

19
34-41
39
20
0
14-15

Based
on
Method
2
0

0

22

0

31
56-67
65
33
0
23-24
                                                                                                            I

                                                                                                            N3

                                                                                                            CO
 * Method 1:. Hood capture efficiency, estimated from linear, regression on fugitive emissions concentrations.
** Method 2: Hood capture efficiency estimated from linear regression on engineering  observer's estimate
             of hood capture efficiency.

-------
                                                TABLE  7.0.5
                                  SUMMARY OF SULFUR DIOXIDE EMISSIONS AND
                                           CONTROL EFFICIENCIES

                                           Ford Motor Company
                                       Rouge Plant - Coke Battery
                                           Dearborn, Michigan
Test
No.
r
1
2
3
4
5
6
7
8
9
Date
(1975)
6/5
6/6
6/6
6/6
*
6/6
6/9
6/9
6/9
6/10
AVERAGE
Average
Gfeen-
.ness
2.0
1.7
1.7
1.7
1.7
2.3
.
2.0
3.0
2.0
Measured
Scrubber
Sulfur Dioxide
Emissions
Inlet
(Ibs/..
ton. of
feed)
0.0084
0.0075
0.0072
0.0091
0.0095
0.0117
0,0091
0.-0058
0.0118
0.0089
Outlet
(Ibs/
ton of
feed)
0.0043
0.0035
0.0032
0.0045
0.0035
0.0094
0.0044
0.0072
0.0074
0.0053
Control
Effi-
ciency
49
53
56
51
63
20
52
0
37
42
Estimated
Average
Hood Capture
Efficiency
(Percent)
Method
. 1*
38
43
43
43
43
33
--
38
22
38
"Method •
2**
61
71
71
71
71
51
--
61
30
61
Estimated
Fugitive
• Emissions •
(Ibs/ton
of feed)
Based
on
Method
I
0.0137
0.0099
0.0095
0.0121
0.0126
0.0238
--
0.0095
0.0418
0.0166
Based
on
Method
2
0.0054
0.0031
0.0029
0.0037
0.0039
0.0112
--
0.0037
0.0275
0.0077
Estimated
Push
Emissions
(Ibs/ton
of feed)
Based
on
Method
I
0.0221
0.0174
0.0167
0.0212
0.0221
0.0355

0.0153
0.0536
0.0255
Baaed
on
Method
2
0.0138
0.0106
0.0101
0.0128
0.0134
0.0229
--
0.0095
0.0393
0.0166
Estimated
Total System
Efficiency
(Percent)
Based
on-
Method
1
19
23
24
22
27
6
-
0
8
16
Based
.00
Method
2
30
38
40
36
45
10
-
0
11
26
                                                                                                            ts>
                                                                                                            to
**Method 2;  Hood capture efficiency estimated from linear regression on engineering observer's
              estimate of hood capture efficiency.

-------
                                               TABLE 7.0-6
                                 SUMMARY OF SULFUR TRIOXIDE EMISSIONS AND
                                          CONTROL EFFICIENCIES

                                          Ford Motor Company
                                       Rouge Plant - Coke Battery
                                          Dearborn, Michigan
Test
No.
1
2
3
4
5
6
7
8
9
Date
(1975)
6/5
6/6
6/6
6/6
*6/6
6/9
6/9
6/9
6/10
AVERAGE
Average
Green-
ness
2.0
1.7
1.7
1.7
1.7
2.3
-
2.0
3.0
2.0
Measured
ScrubbjBj:
Sulfur Trioxide
Emissions
Inlet
(Ibs/.-
ton of
feed)
CO. 00043
0.00084
<0. 00043
C0.00045
<0.00043
<0.0018
<0.00072
0.0088
<0. 00070
TSBlT
Outlet
(Ibs/
ton of
feed)
<0. 00059
<0. 00062
<0. 00058
<0.0013
0.0025
<0.0015
<0. 00076
<0.00079
<0.00089
0.0003 —
0.00 1 1
Control
Effi- .
eiency
(t)
-
>26
-
-
0
-
-
>91

39-67
Estimated
Average
Hood Capture
Efficiency
(Percent)
Method
1*
38
43
43
43
43
33
-
38
22
38
'Method •
2**
61
71
71
71
71
51
-
61
30
61
Estimated
Fugitive
Emissions •
(Ibs/ton
of feed)
Based
on
Method
1
< 0.000 70
0.0011
C 0.00057
C 0.00060
<0.00057
<0.0037
--
0.014
< 0.002 5
0.0019-
0.0030
Based
on
Mejftiod
2
<0.00027
<0. 00034
<0.00018
< 0.00018

-------
                       TABLE  7.0-7
SUMMARY OF TOTAL SULFUR -OXIDES- (AS SULFATE) EMISSIONS AND
                  CONTROL EFFICIENCIES
                   Ford Motor Company
               Rouge Plant  -  Coke Battery
                   Dearborn,  Michigan
Test
No.
1
2
3
4
5
6
Date
(1975)
6/3
6/4
6/4
6/4
*6/4
6/5
AVERAGE
Average
Green-
ness
2.3
1.3
2.0
1.0
3.0
3.0
2.:i
Measured
Scrubber
Total Sulfur. Oxides
(As Sulfate)
Emissions
Inlet
(Ibs/.-
ton of
feed)
0.229
0.6856
0.0353
0.0416
0.0587
0.0563
0.0844
Outlet
(Ibs/
ton of
feed)
0.0588
0.0170
0.0120
0.0207
0.0343
0.0244
0.0279
Control
Effi-
ciency
(X)
74.3
80,1
66.0
50.2
41.6
56.7
61.5
Estimated
Average
Hood Capture
Efficiency
(Percent)
Method
"... 1*
33
49
38
54
22
22
36
'Method •
2**
51
82
61
92
30
30
58
Estimated
Fugitive
. Emissions
(IBS/ton
of feed)
Based •'
on
Method
1
0.465
0.0891
0.0576
0.0354
0.208
0.200
0.176
Baaed .
on
Method
2
0.220
0.0188
0.0226
0.0036
0.137
0.131
0.089
Estimaceo,
Push
Emissions
libs/ton
of feed)
Based
on
Method
1
0.694
0.175
0.0929
0.0770
0.267
0.256
0.260
Based
on
Method
2
0.449
0.104
0.0579
0.0452
0.196
0.187
0.173
Estimated
Total System >
Efficiency
(Percent)
Based
on-
Method
1
25
.39
25
27
9
12
23
Baaed
.•0
Method
2
38
66
40 '
46
: 13
17
37
*Method 1: Hood capture efficiency estimated from linear regression on fugitive emissions oncenfc atio'na
**Method 2; Hood capture efficiency estimated from linear regression on engineering observer's
estimate of hood capture efficiency.

-------
                                               TABLE  7.0-8
                                  SUMMARY OF CARBON MONOXIDE EMISSIONS AND
                                          CONTROL EFFICIENCIES
                                           Ford Motor Company
                                       Rouge Plant -  Coke Battery
                                           Dearborn,  Michigan
Test
No.
2
5
8
11
13
16
Date
(1975)
6/4
6/4
6/4
6/12
*7/16
7/16
AVERAGE
Average
Green-
ness
3
1
-
1
2
1
1.6
Measured
Scrubber
Carbon Monoxide
Emissions
Inlet
(Ibs/..
ton of
feed)
0.021
00.0032
<0.0033
<0.0033
0.034
0.038
0. 016-
0.017
Outlet
(Ibs/
ton of
feed)
<0.0034
<0.0035
<0.0036
<0.0036
<0.022
<0.022
<0.010
Control
Effi-
ciency
(X)
>84
-
-
-
>35
>42
>54
Estimated
Average
Hood Capture
Efficiency
(Percent)
Method
I*
22
54
-
54
38
54
44
Method-
2**
30
92
'-
92
61
92
73
Estimated
Fugitive
Emissions
(Ibs/ton
of feed)
Based
on
Method
1
0.074
<0.0027
--
<0.0028
0.055
0.032
0.033
Based
on
Mejfhod
2
0.049
<0.00028
--
<0j00029
0.022
0.0033
0.015
Estimated
Push
Emissions
(Ibs/ton
of feed)
Based
on
Method
1
0.095
<0.0059
--
<0.0061
0.089
0.070
0.051 -
0.053
Based
on
Method
2
0.070
<0.0035
—
<0.0036
0.056
0.041
0.033-
0.035
Estimated
Total System
Efficiency
(Percent)
Based
on-
Method
1
19-22
,
-
-
13-38
23-54
18-38
Based
.on
Method
2
25-30
-
-
-
21-61
38-92
28-61
                                                                                                            NJ
                                                                                                            IS)
 *Method
**Method 2:
Hood capture efficiency estimated from linear regression on  engineering observer's
 estimate of hood capture efficiency.

-------
                   TABLE  7.0-9
SUMMARY OF NITROGEN OXIDES (as N02) EMISSIONS AND
              CONTROL EFFICIENCIES
              Ford Motor  Company
           Rouge  Plant  - Coke  Battery
              Dearborn, Michigan
Test
No.



1
2
3
4

5
6
7
8
9
10
11
12
13
Date
(1975)



6/10
6/10
6/10
6/10

6/10
6i/ll
6/11
6/11
6/11
7/18
7/18
7/18
7/18

AVERAGE
Average
Green-
ness



1
3
1
1

3
1
2
1
1
1
1
2
3

1.6
Measured
Scrubber
Nitoogen Oxides
Emissions
Inlet
(Ibs/.
ton of
feed)
0.0064
0.0170
0.0050
0.0033

0.0080
0.0018
0.0139
<0.0017
0.0017
<0.0017
0.0017
0.0017
0.0017
0.0048-
0.0050
Outlet
(Ibs/
ton of
feed)
0.0118
0.0205
0.0051
0.0051

0.0395
0.0050
0.0281
0.0016
0.0017
<0.0018
0.0019
0.0035
0.0124
0.0105-
0.0106
Control
Effl- -
ciency
(1)
0
0
0
0

0
0
0
0
0
—
e
0
0

0
Estimated
Average
Hood Capture
Efficiency
(Percent)
Method
1*


54
22
54
54

22
54
38
54
54
54
54
38
22

44
'Method •
2**


92
30
92
92

30
92
61
92
92
92
92
61
30

73
• Estimated
Fugitive
Emissions
(Ibs/ton
of feed)
Based
on
Method
1
0.0055
0.0603
0.0043
0.0028

0.028
0.0015
0.0227
0.0014
0.0014

00
..










•
**Method 2: Hood capture efficiency estimated from linear regression on engineering observer's
estimate of hood capture efficiency.

-------
                                               TABLE 7.0-10
                                     SUMMARY OF BENZENE.EMISSIONS AND
                                          CONTROL EFFICIENCIES
                                           Ford Motor Company
                                       Rouge Plant - Coke Battery
                                           Dearborn, Michigan
Test
No.
1
3
5
7
9
11
Date
(1975)
6/5
6/5
6/5
6/5
'6/6
6/6
AVERAGE
Average
Green-
ness
3
3
2
2
1.7
2.3
2.3
Measured
Scrubber
Benzene
Emissions
Inlet
(Ibs/,
ton of
fee'd)
0.0007
0.002
<0. 0002
< 0.0002
£0.00008
C0.00008
0.0004-
0.0005
Outlet
(Ibs/
ton of
feed)
<0.0003
<0.0003
<0.0003
<0. 0003
<0.0001
< 0.0001
<0.000 2
Control
Effi-
ciency
(I)
>57
>85
• -
-
-


Estimated
Average
Hood Capture
Efficiency
Method
I*
22
22
38
38
43
33
33
Method-
2**
30
30
61
61
71
51
51
Estimated
Fugitive
Emissions
(Ibs/ton
of feed)
Based
on
Method
1
0.002
0.007
<0.0003
<0.0003
CO. 0001
C0.0002
0.002
Based
on
Mejttiod
2
0.002
0.005
oo8
0.001
Estimated
Push
Emissions
(Ibs/ton
of feed)
Based
on
Method
1
0.003
0.009
<0.0005
<0.0005
< 0.000 2
< 0.00 03
•MMMOMMIM^BM
0.002
Baaed
on
Method
2
0.003
0.007
<0.0003
<0.0003
< 00)0011
< 000016
0.002
Estimated
Total System
Efficiency
(Percent)
Based
on
Method
1
23-33
19-22
-
-


21-28
Based
.on
Method
2
23-33
24-29
-
-
-

24-31
**Method 2;  Hood capture efficiency estimated from linear regression on engineering  observer's
              estimate of hood capture efficiency.

-------
                                               TABLE  7.0-11
                                SUMMARY OF BENZENE & HOMOLOGUES EMISSIONS ANP
                                          CONTROL EFFICIENCIES"
                                           Ford Motor Company
                                       Rouge Plant - Coke Battery
                                           Dearborn, Michigan
Test
No.



1
3
5
7
9
11
Date
(1975)



6/5
6/5
6/5
6/5
6/6
6/6
AVERAGE
Average
Green-
ness



3
3
2
2
1.7
2.3
2. .3
Measured
Scrubber
Benzene & Homologues
Emissions
Inlet
(Ibs/..
ton- of
feed)
0.0012
0.0372
0.0024
0.00070
0.00079
0.0050
0.0079
Outlet
(Ibs/
ton of
feed)
0.0076
0.0105
0.0016
0.0022
0.00073
0.00031
0.0038
Control
Effi-
ciency
(%)
0
72
33
0
8
94
34
Estimated
Average
Hood Capture
Efficiency
(Percent)
Method
1*


22
22
38
.38
43
33
33
'Method-
2**


30
30
61
61
71
51
51
Estimated
Fugitive
Emissions
Based
on
Method
1
0,0043
0.132
0.0039
0.0011
0.0010
0.010
0.025
Based
on
Metho'd
2
0.0028
0.0868
0.0015
0.00045
0.00032
0.0048
0.016
Estimated
Push
Emissions
Based
on
Method
1
0.0055
0.169
0.0063
0.0018
0.0018
0.015
0.033
Based
on
Method
2
0.0040
0.1240
0.0039
0.00115
0.00111
0.0098
0.024
•Estimated
Total System
Efficiency
(Percent)
Based
on-
Method
1
. 0
16
13
0
4
31
11
Based
,en
Method
2
0
22
21
0
5
48
16







.
to
10
O
1


 *Method 1:
**Method 2;
Hood capture efficiency estimated from linear regression on engineering observer's
 estimate of hood capture efficiency.

-------
                                               TABLE 7.0-12
                        SUMMARY OF TOTAL LIGHT HYDROCARBONS (AS METHANE) EMISSIONS AND
                                          CONTROL EFFICIENCIES
                                           Ford Motor Company
                                       Rouge Plant - Coke Battery
                                           Dearborn, Michigan
Test
No.
1
4
7
10
14
17

Date
(1975)
6/4
6/4
6/4
6/12
7/16
7/16

AVERAGE
Average
Green-
ness
2
1
3
1
1
1

1.5
Measured
Scrubber
Total Light Hydrocarbons
Emissions
Inlet
(Ibs/..
ton of
feed)
0.0073
0.0056
0.0008
0.0006
0.001
0.0968

D.019
Outlet
(Ibs/
ton of
feed)
0.012
0.0034
0.0053
0.0006
0.0027
0.0022

0.0044
Control
Effi-
ciency
ftt
0
39
0
0
0
98

23
Estimated
Average
Hood Capture
7ff£cin??y
Method
1*
38
54
22
54
54
54

46
'Method-
2**
61
92
30
92
92
92

76
Estimated
Fugitive
Emissions
(of"re13)
Based
oa
Method
1
0.012
0.0048
0.003
0.0005
0.0009
0.0825
t
0.017
Based
on
Mejehod
2
0.0047
0.00049
0.002
0.00005
0.00009
a 00 84 2

0.0026
Estimated
Push
Emissions
ciHss»>
Based
on
Method
1
0.019
0.0104
0.004
0.0011
0.002
0.1793

0.036
Baeed
on
Method
0.0120'
0.0061
0.003
0.0006
0.001
0.1052

0.021
Estimated
Total System
Efficiency
(Percent)
Based
en-
Method
1
0
21
0
0
0
53

12
Based
.en
Method
2
0
36
0
0
0
90

21
**Method 2:
Hood capture efficiency estimated from linear regression on engineering  observer's
 estimate of hood capture efficiency.

-------
                                               TABLE 7.0-13
                                 SUMMARY OF METHANE  & HOMOLOGUES EMISSIONS AND
                                           CONTROL EFFICIENCIES

                                           Ford Motor Company
                                       Rouge Plant - Coke Battery
                                           Dearborn, Michigan
Test
No.
1
4
7
10
14
17

Date
(1975)
6/4
6/4
6/4
6/12
'7/16
7/16

AVERAGE
Average
Green-
ness
2
1
3
1
1
1

1.5
Measured
Scrubber
Methane & Homologues
Emissions
Inlet
(Ibs/..
ton of
feed)
0.001
0.003
0.0004
0.0004
0.0006
0.0004

D.001
Outlet
(Ibs/
ton of
feed)
0.003
0.001
0.0006
0.0006
0.0004
0.0006

0.001
Control
Effi-
ciency
0
67
0
0
33
0

17
Estimated
: Average
Hood Capture
Efficiency
(Percent)
Method
1*
38
54
22
54
54
54

46
Method •
2**
61
92
30
92
92
92

76
• Estimated
Fugitive
Emissions
C-l> D fl / t OQ
o £ £Q G d )
Based
on
Method
1
0.002
0.003
0.001
0.0003
0.0005
0.0003
1
0.001
Based
on
MeJThod
2
0.0006
0.0003
0.0009
0.00003
OJ00005
0.00003

0.0003
Estimated
; Push
Emissions
C 1 tt 8 L. t OQ
o £ * 6 6 d j
Based
on
Method
1
0.003
O.OQ6
0.001
0.0007
0.0011
0.0007

0.002
Based
on
Method
2
0.002 '
0.003
0.0013
0.0004
0.0006
0.0004

0.001
Estimated
Total System
Efficiency
(Percent)
Based
on-
Method
1
0
33
0
0
18
0

8
Based
.on
Method
2
0
57
0
0
25
0

14


i
10
NJ
1


 *Method
**Method 2:
Hood capture efficiency estimated from linear regression on engineering  observer's
 estimate of hood capture efficiency.

-------
                    TABLE  7.0-14
SUMMARY  OF ETHYLENE  & HOMOLOGUES  EMISSIONS  AND
                CONTROL EFFICIENCIES
                 Ford Motor Company
             Rouge Plant -  Coke Battery
                 Dearborn,  Michigan
Test
No.



1
4

7


10
14
17
Date
(1975)



6/4
6/4

6/4


6/12
7/16
7/16
AVERAGE
Average
Green-
ness



2
1

3


1
1
1
1.5
Measured
Scrubber
Ethylene & Homologues
Emissions
Inlet
(Ibs/.
ton of
feed)
0.001
0.005

0.0004


CO. 0002
£0.0002
0.0606
0.011
Outlet
(Ibs/
ton of
feed)
0.006
0.0003

0.0007


<0.0002
0.0007
0.0002
0.001
Control
Effi-
ciency
m
0
94

0


-
0
99.7
39
Estimated
Average
Hood Capture
Efficiency
(Percent;
Method
1*


38
54

22


54
54
54
46
Method •
2**


61
92

30


92
92
92
76
Estimated
Fugitive
Emissions
(obf8/fl^)
Based
on
Method
1
0.002
0.004

0.001


CO. 0002
: o.oo 02
0.0516
0.010
Based
on
Method
2
0.0006
0.0004

a 0009



-------
                        -  234  -


                    TABLE  7.3.1-1

             PARTICLE SIZE DISTRIBUTION

                 Ford Motor Company
             Rouge Plant - Coke Battery
                 Dearborn, Michigan



Test Number:   Inlet  1  (A,  Ba  and  C)	
Characteristic
Diameter of
Particles
(M)
>7.3
7.3 - 4.9
4.9 - 3.5
3.5 - 2.1
2.1 - 1.1
1.1 - 0.67
0.67- 0.46
<0.46
Weight
(mg)
421.2
5.5
7.6
6.0
7.2
3.7
1.7
1.4
Size Distribution
by Weight
Percent
92.7
1.2
1.7
1.3
1.6
0.8
0.4
0.3
Cumulative
Percent
100.0
7.3
6.1
4.4
3.1
1.5
0.7
0.3
                     Clayton Environmental ConsuLtants,  Inc,

-------
                        - 235 -


                    TABLE 7.3.1-2

             PARTICLE SIZE  DISTRIBUTION

                 Ford Motor  Company
             Rouge Plant -  Coke Battery
                 Dearborn, Michigan
Test Number:
Inlet 2 (A. B, and C)
Characteristic
Diameter of
Particles
(n)
>7.2
7.2 - 4.7
4.7 - 3.3
3.3 - 2.1
2.1 - 1.1
1.1 - 0.65
0.65- 0.44
<0 . 44
Weight
(mg)
359.9
6.9
6.5
7.2
8.6
4.7
2.0
1.0
Size Distribution
by Weight
Percent
90.7
1.7
1.6
1.8
2.2
1.2
0.5
0.3
Cumula tive
Percent
100.0
9.3
7.6
6.0
4.2
2.0
0.8
0.3
                     Clayton Environmental Consultants, Inc.

-------
                        - 236 -


                    TABLE  7.3.1-3

             PARTICLE SIZE DISTRIBUTION

                 Ford Motor  Company
             Rouge Plant - Coke Battery
                 Dearborn, Michigan
Test Number:
Inlet 3 (A, B, and C)
Characteristic
Diameter of
Particles
<")
>7.2
7.2 - 4.8
4.8 - 3.3
3.3 - 2.1
2.1 - 1.1
1.1 - 0.66
0.66- 0.45
<0.45
Weight
(mg)
517.2
7.9
16.2
5.2
9.2
4.4
18.3
1.2
Size Distribution
by Weight
Percent
89.2
1.4
2.8
0.9
1.6
0.7
3.2
0.2
Cumula tive
Percent
100.0
10.8
9.4
6.6
5.7
4.1
3.4
0.2
                     Clayton Environmental Consultants,  Inc,

-------
                        -  237  -



                    TABLE' 7.3.1-4

             PARTICLE SIZE DISTRIBUTION

                  Ford Motor Company
             Rouge  Plant  - Coke  Battery
                  Dearborn, Michigan
Test Number:
Inlet 4 (A, B, and C)
Characteristic
Diameter of
Particles
7.4
7.4 - 4.9
4.9 - 3.4
3.4 - 2.1
2.1 - 1.1
1.1 - 0.67
0.67- 0.46
<0.46
Weight
(mg)
301.5
6.8
6.6
9.0
11.7
8.1
4.6
2.2
Size Distribution
by Weight
Percent
86.0
2.0
1.9
2.6
3.3
2.3
1.3
0.6
Cumula tive
Percent
100.0
14.0
12.0
10.1
7.5
4.2
1.9
0.6
                     Clayton Environmental Consultants, Inc.

-------
                        -  238  -


                    TABLE  7.3.1-5    ^

             PARTICLE SIZE DISTRIBUTION

                 Ford Motor Company
             Rouge Plant - Coke Battery
                 Dearborn, Michigan
Test Number:
Inlet 5 (A)
Characteristic
Diameter of
Particles
(u)
>7 . 1
7.1 - 4.7
4.7 - 3.3
3.3 - 2.1
2.1 - 1.1
1.1 - 0.65
0.65- 0.44
<0.44
Weight
(mg)
76.1
1.7
1.3
1.6
1.6
1.0
0.5
<0.5
Size Distribution
by Weight
Percent
90. 3
2.0
1. 5
1.9 .
1.9
1.2
0.6
<0.6
Cumulative
Percent
100.0
9.7
7. 7
6.2
4.3
2.4
1.2
0.6
                     Clayton  Environmental  Consultants,  Inc,

-------
                        - 239 -



                    TABLE 7 .3.1-6    ^

             PARTICLE SIZE  DISTRIBUTION

                 Ford Motor Company
             Rouge Plant -  Coke Battery
                 Dearborn, Michigan
Test Number:
Inlet 6 (A, B, and C)
Characteristic
Diameter of
Particles
(.H)
>7.1
7.1 - 4.8
4.8 - 3.3
3.3 - 2.1
2.1 - 1.1
1.1 - 0.65
0.65- 0.45
<0.45
Weight
(ing)
250.3
6.8
6.0
8.3
7.4
4.7
1.5
0.8
Size Distribution
by Weight
Percent
87.6
2.4
2.1
2.9
2.6
1.6
0.5
0.3
Cumula tive
Percent
100.0
12.4
10.0
7.9
5.0
2.4
0.8
0.3
                     Clayton Environmental Consultants, Inc.

-------
                        - 240 -


                    TABLE  7.3.1-7

             PARTICLE SIZE DISTRIBUTION

                 Ford Motor  Company
             Rouge Plant - Coke Battery
                 Dearborn, Michigan



Test Number:  Inlet 7 (A,  B, and C)	
Characteristic
Diameter of
Particles
7.2
7.2 - 4.8
4.8 - 3.4
3.4 - 2..1
2.1 - 1.1
1.1 - 0.66
0.66- 0.45
<0.45
Weight
(og)
283.9
6.8
3.7
6.1
5.5
2.3
0.7
<0.5
Size Distribution
by Weight
Percent
91. 7
2.2
1. 2
2.0
1.8
0.7
0.2
<0.2
Cumulative
Percent
100.0
8. 3
6.1
4. 9
2. 9
1.1
0.4
0 .2
                     Clayton Environmental Consultants,  Inc,

-------
                        -  241 -
                     TABLE 7.3.1-8

             PARTICLE SIZE DISTRIBUTION

                 Ford Motor Company
             Rouge Plant - Coke Battery
                 Dearborn, Michigan
Test Number:   Outlet  1
Characteristic
Diameter of
Particles
(n)
>9.0
9.0 - 6.0
6.0 - 4.1
4.1 - 2,6
2.6 - 1.4
1.4 - 0.81
0.81- 0.57
<0.57
Weight
(mg)
3.9
0.8
0.7
0.4
3.6
0.8
1.1
0.4
Size Distribution
by Weight
Percent
33.3
6.8
6.0
3.4
30.8
6.9
9.4
3.4
Cumula tive
Percent
100.0
66.7
59.9
53.9
50.5
19.7
12.8
3.4
                     Clayton Environmental Consultants, Inc,

-------
                        -  242  -


                     TABLE 7.3.1-9    ^

             PARTICLE SIZE DISTRIBUTION

                 Ford Motor Company
             Rouge Plant  - Coke Battery
                 Dearborn, Michigan
Test Number:
Outlet 2
Characteristic
Diameter of
Particles
7.1
7.1 - 4.7
4.7 - 3.4
3.4 - 2.1
2.1 - 1.1
1.1 - 0.65
0.65- 0.44
<0.44
Weight
(mg)
1.3
<0.5
<0.5
<0.5
<0.5
<0.5
<0.5
0.9
Size Distribution
by Weight
Percent
25.0
<9.6
<9.6
<9.6
<9.6
<9.6
<9.6
17.4
Cumulative
Percent
100.0
75.0
65.4
55.8
46.2
36.6
27.0
17.4
                     Clayton Environmental  Consultants,  Inc.

-------
                        -  243  -


                    TABLE  7.3.1-10

             PARTICLE SIZE DISTRIBUTION

                 Ford Motor Company
             Rouge Plant - Coke Battery
                 Dearborn, Michigan
Test Number:  Outlet 3
Characteristic
Diameter of
Particles
6.4
6.4 - 4.2
4.2 - 2.9
2.9 - 1.9
1.9 - 0.94
0.94- 0.58
0.58- 0.39
<0.39
Weight
(mg)
2.0
0.6
0.6
<0.5
0.8
1.5
1.0
0.6
Size Distribution
by Weight
Percent
26.3
7.9
7.9
<6.6
10.5
19.7
13.2
7.9
Cumulative
Percent
100.0
73.7
65.8
57.9
51.3
40.8
21.1
7.9
                     Clayton Environmental Consultants, Inc.

-------
                        - 244 -
                    TABLE 7.3.1-11

             PARTICLE SIZE DISTRIBUTION

                 Ford Motor Company
             Rouge Plant - Coke Battery
                 Dearborn, Michigan
Test Number:
Outlet 4
Characteristic
Diameter of
Particles
(u)
>6.4
6.4 - 4.2
4.2 - 2.9
2.9 - 1.8
1.8 - 0.92
0.92- 0.56
0.56-- 0.38
<0.38
Weight
(«ng)
<0.5
<0.5
<0.5
<0.5
<0.5
<0.5
0.5
0.9
Size Distribution
by Weight
Percent
<11.4
<11.4
<11.4
<11.4
<1T.3
<11.3
11.3
20.5
Cumulative
Percent
100.0
88.6
77.2
65.8
54.4
43.1
31.8
20.5
                     Clayton Environmental Consultants,  Inc,

-------
                        -  245  -



                    TABLE  7.3.1-12   >

             PARTICLE SIZE DISTRIBUTION

                 Ford Motor Company
             Rouge Plant - Coke Battery
                 Dearborn, Michigan
Test Number:
Outlet 5
Characteristic
Diameter of
Particles
(n)
>6.2
6.2 - 4.1
4.1 - 2.9
2.9 - 1.8
1.8 - 0.91
0.91- 0.56
0.56- 0.38
<0.38
Weight
(mg)
<0.5
<0.5
<0.5
<0.5
<0.5
<0.5
<0.5
<0.5
Size Distribution
by Weight
Percent
	
	
	
	
	
	
	

Cumulative
Percent
	
	
	
	
	
	
	

                     Clayton Environmental Consultants, Inc,

-------
Test Number:
           -  246  -


       TABLE' 7.3.1-13

PARTICLE SIZE DISTRIBUTION

    Ford Motor Company
Rouge Plant - Coke Battery
    Dearborn, Michigan



   Outlet 6
Characteristic
Diameter of
Particles '
(n>
>6.1
6.1 - 4.1
4.1'- 2.9
2.9 - 1.8
1.8 - 0.91
0.91 - 0.55
0.55 - 0.38
<0.38
Weight
(mg)
<0.5
<0.5
<0.5
0.6
<0.5
0.6
<0.5
2.6
Size Distribution
by Weight
Percent
< 7.9
< 7.9
< 7.9
9.6
< 7.9
9.6
< 7.9
41.3
Cumulative
Percent
100.0 '
92.1
84.2
76.3
66.7
58.8
49.2
41.3
                     Clayton Environmental Consultants, Inc,

-------
                        - 247 -


                    TABLE  7.3.1-14

             PARTICLE SIZE  DISTRIBUTION

                  Ford Motor Company
             Rouge  Plant -  Coke  Battery
                  Dearborn,  Michigan
Test Number:
Outlet 7
Characteristic
Diameter of
Particles
(n)
>6.3
6.3 - 4.2
4.2 - 3.0
3.0-1.9
1.9 - 0.93
0.93- 0.57
0.57- 0.39
<0.39
Weight
(mg)
0.9
<0.5
<0.5
<0.5
<0.5
<0.5
<0.5
<0.5
Size Distribution
by Weight
Percent
20.5
<11.4
<11.4
<11.4
<11.4
<11.3
<11.3
<11.3
Cumulative
Percent
100.0
79.5
68.1
56.7
45.3
33.9
22.6
11.3
                     Clayton Environmental  Consultants,  Inc,

-------
Effective
Particle
Diameter
(microns)

                                            FIGURE 7.3.1-1
                                      PARTICLE SIZE DISTRIBUTION
                                            SCRUBBER INLET
                                          Ford Motor Company
                                     Rouge Plant - Coke Battery
                                         Dearborn, Michigan
                                                                                                          I

                                                                                                          Ni

                                                                                                          00

                                                                                                          I
                                      I    10    t»   K   40  H  ta

                                        Cumulative  Percentage

-------
Effective
Particle
Diameter
(microns)
                                                Igt IB* m. l *; Ml M-
                                             FIGURE  7.3.1-2
                                       PARTICLE  SIZE  DISTRIBUTION
                                            SCRUBBER OUTLET
                                           Ford_Mot:or  Company
                                       Rouge Plant -:  CoEe~Slttery
                                           Dearborn, Michigan
                                                                                                        I

                                                                                                        NJ

                                                                                                        VO
                                      S    18     20   10  40   50  10

                                        Cumulative Percentage

-------
                 - 250 -
             TABLE 7.3.1-15

   CONCENTRATION OF PARTICIPATE MATTER
CALCULATED FROM" PARTICLE SIZING SAMPLES

          Ford  Motor  Company
     Rouge  Plant  - Coke Battery
          Dearborn, Michigan
Sampling
Station
Coke Battery A
Scrubber
Inlet


















Coke Battery A
Scrubber
Outlet










Test
No.
1A
IB
1C
2A
2B
2C
3A

3B
3C
4A
4B
4C
5A
6A
6B

6C
7A
7B
7C
1

2

3

4

5
6

7

1975
Date
9/4
9/4
9/5
9/5
9/8
9/8
9/9
9/10
9/10
9/10
9/10
9/11
9/11
9/11
9/12
9/12
9/15
9/15
9/15
9/15
9/16
9/4
9/5
9/5
9/6
9/9
9/10
9/10
9/11
9/11
9/12
9/15
9/15
9/16
Sampling
Period
Start
09:00
14:25
09:04
12:51
10:36
13:37
09:59
08:22
10:05
11:55
14:06
08:41
10:41
14:06
08:28
12:25
08:25
09:55
12:23
14:25
08:34
09:00
09:04
12:51
10:36
09:59
08:22
14:06
08:41
14:06
09:56
08:25
12: 23
08:34
Stop
09:52
15: 14
09:31
13:32
11:52
14;32
10:00
08:36
10:35
12:26
14:28
09: 10
11:19
14:26
10:10
12:39
08:26
10:11
13:23
14:57
08:56
15: 14
09:31
13:32
14:32
10:00
12:26
14:28
11:19
14:26
12:39
10: 11
14:57
08:56
Sampled
Volume
(DSCF)

6.00


6.78



6.65


6.77

2.36


7.00


7.04

4.26

7.12

9.34

9.51

3.28
9.62

9.11

Sample
Weight
:(mg)

454.3


396.8



579.6


350.5

84.3


285.8


309.5

11.7

2.2

7.1

1.4

<0.5
3.8

0.9

Concentration
(gr/DSCF)

1.17


0.903



1.34


0.799

0.551


0.630


0.678

0.042

0.005

0.012

0.002

<0.002
0.006

0.002


-------
              TABLE 7.3.1-16

COMPOSITION OF PARTICULATE MATTER, PERCENT
              SCRUBBER INLET

            Ford Motor Company
        Rouge Plant - Coke Battery
            Dearborn, Michigan
i

N>
Test
No.
1
2
3
4
5
6
7
8
9
Sulfate
Filterable
0.3
0.4
0.4
0.4
0.3
0.5
0.5
0.5
0.6
Total
1.5
1.2
1.8
3.4
1.3
1.8
1.6
1.0
0.9
Iron
Filterable
1.4
1.1
1.6
0.9
0.6
0.4
0.5
0.4
0.5
Total
1.4
1.1
1.5
0.8
0.6
0.4
0.5
0.4
0.5
Aluminum
Filterable
0.7
0.9
0.6
0.6
0.5
0.6
0.3
0.2
0.3
Total
0.7
0.9
0.6
0.5
0.5
0.6
0.2
0.2
0.3
Titanium
Filterable
<0.05
<0.06
<0.05
<0.1
—
—
—
—
—
Total
<0.08
<0.1
<0.08
<0.1
—
—
—
—
—
l
Chloride
Filterable
0.2
0.2
0.2
0.3
0.2
0.2
0.09
0.1
0.06
Total
0.2
0.2
0.2
0.3
0.2
0.2
0.09
0.1
0.06
                               Clayton Environmental  Consultants,  Inc.

-------
              TABLE 7.3.1-17
COMPOSITION OF PARTICULATE MATTER, PERCENT
              SCRUBBER  OUTLET

            Ford Motor Company
        Rouge Plant - Coke Battery
            Dearborn, Michigan
ro
Ol
10
Test
No.
I
2
3
4
5
6
7
8
9
Sulfate
Filterable
<38.4
66.2
33.8
50.6
<29.8
<28.7
46.0
61.7
45.9
T.Ota 1
16.2
27.0
30.0
30.6
19.8
26.7
44.2
55.6
31.9
Iron
Filterable
2.2
0.5
1.2
2.9
0.9
<0.05
1.1
1.6
0.9
Total
0.5
0.1
0.3
0.7
0.6
0.1
1.0
1.5
0.4
Aluminum
Filterable
2.9
<0.7
1.4
<1.3
6.9
1.5
11.3
6.4
3.0
Total
0.7
<0.4
0.4
<0.6
3.6
0.7
7.2
3.3
1.1
Titanium
Filterable
<3.6
<4.2
<4.0
<7.6
—
—
—
—
—
Total
<1.8
<2.3
<2.0
<3.8
—
—
—
—
—
Chloride
Filterable
9.8
24.2
31.6
10.6
3.7
4.9
2.0
5.5
6.2
Total
2.4
6.7
9.6
2.6
1.9
2.3
4.9
2.9
2.3
                               Clayton Environmental  Consultants, Inc.

-------
                                  -  253  -


                       - 0    TABLE 7.3.9-1

                 SUMMARY OF TOTAL UNCONTROLLED EMISSIONS

                           Ford Motor Company
                       Rouge Plant  - Coke Battery
                           Dearborn, Michigan
               Contaminant
                                            Total Uncontrol
                                                  (Ibs/ton
                led Emissions
                of feed)
                                                Range*
                   Average
                                                                     **
  Filterable  Particulate

  Condensibles

  Sulfur Dioxide

 Sulfur Trioxide

 Total Sulfur Oxides (as Sulfate)

 Carbon Monoxide

 Nitrogen Oxides (as N02)

 Benzene

 Benzene and Homologues

 Total Light Hydrocarbons (as Methane)

 Methane and Homologues

 Ethylene and Homologues
 1.23  - 3.76

 0.0116- 0.144

 0.006 - 0.081

<0.0008- 0.014

 0.024 - 0.52

<0.004 - 0.074

<0.002 - 0.081

<0.0001- 0.007

 0.001 - 0.14

 0.006 - 0.085

 0.0004- 0.005

<0.0002- 0.052
   2.28

   0.067

   0.018

0.002-0.003

   0.16

0.024-0.026

   0.019

0.001-0.002

   0.025

   0.014

   0.002

   0.007
*   The range is defined by the lowest and highest values in each respec-
    tive Table 7.0-1 to 7.0-14 without regard for the method of hood
    capture efficiency estimation.

**  The average is the arithmetic average of all values in each respec-
    tive Table 7.0-1 to'7.0-14 without regard for the method of hood
    capture efficiency estimation.

-------
                                             TABLE  7.3.9-2
                        SUMMARY OF MEASURED EMISSIONS FOR CONTAMINANTS  PRESENT
                                     BELOW LIMITS OF DETECTABILITY

                                          Ford Motor Company
                                      Rouge Plant - Coke Battery
                                          Dearborn, Michigan
                                                                           to
                                                                           Ul
                                                                           -p-
      Contaminant
                             Scrubber  la-let  (Ibs/ton  of  feed)
                                Range
                         Average
                                           Scrubber Outlet  (Ibs/ton  of feed)
                       Range
                          Average
Acetylene
Acrid ine
Filterable Acrolein
Total Acrolein
Ammonia
Anthracene
Benzo(a+e)pyrene
Carbazole
Carbon Bisulfide
Chlorine
Chrysene
Filterable Cyanide
Total Cyanide
Fluoranthene
Fluorene
9-Fluorenone
Hydrogen Sulfide
Filterable Naphthalene
Total Naphthalene
Phenanthrene
Filterable Phenolics
Total Phenolics
Pyrene
Pyridine
<0.00015-<0v00017
<0.00008-<0.00009
<0.00008-<0.00009
<0.00004-<0.00005
<0.00008-<0.00009
<0.004  -<0.011

<0.00004-<0.00005
,02
.00004
.00004
,00004
.42
.00009
.00004
.00008
.00055
.00098
.00004
.00008
        • 0.43)*
        -<0.00005
        •<0.00005
        -<0.00005
        •<0.43)*
         <0.00010
        -<0.00005
        •<0.00009
        -<0.0011
        -<0.0016
        -<0.00005
         <0.0002
  <0.00016
  <0.00008
  <0. 00008
  <0.00004
  <0.00008
  <0.008

  <0.00004

(0.14-0.16)*
  <0.00004
  <0.00004
  <0.00004
 (<0.42)*
  <0.00010
  <0.00004
  <0.00008
  <0. 00069
  <0.0011
  <0.00004
 X0.0002
<0.00016 -<0.00019
<0.00007 -<0.00009
<0.000086-<0.000096
<0.00037 -<0.00035
<0.005   -<0.004
<0.00007 -<0.00009
     <0.00004
<0.00007 -<0.00009
<0.005   -<0.015
     <0.000009
     <0.00004
 0.00010 - 0.00019
 0.00065 - 0.00069
     <0.00004
     <0.00004
     <0.00004
     <0.001

<0.00004- 0.0029
<0.00007-<0.00009
<0.00042-<0.00047
<0.00080-<0.00086
     <0.00004
<0.0001 -<0.0003
    <0.00018
    <0.00008
    <0.000091
    <0.00036
    <0.004
    <0. 00008
    <0.00004
    <0.00008
    <0.010
    <0.000009
    <0.00004
     0.00014
     0.00067
    <0.00004
    <0.00004
    <0.00004
    <0.001

0.00058-0.00068
    <0.00008
    <0. 00044
    <0.00082
    <0.00004
    <0.0002
  * Concentration in ppm due to  lack of  coal  charge  data.
                                                             Clayton  Environmental  Consultants, Inc.

-------
                   - 255 -
              FIGURE 8.2.1-1
A CLEAN PUSH WITH CONTROL  SYSTEM  OPERATING
            Ford Motor  Company
        Rouge Plant - Coke Battery
            Dearborn, Michigan
                                                  30-29
                                                   31-28

-------
                   - 256 -

        FIGURE  8.2.1-1  (continued)

A CLEAN PUSH WITH CONTROL  SYSTEM OPERATING

            Ford Motor  Company
        Rouge Plant  - Coke  Battery
            Dearborn, Michigan
                                                   30-30
                                                   31-29

-------
                   - 257 -
        FIGURE 8.2.1-1  (continued)

A CLEAN PUSH WITH CONTROL SYSTEM  OPERATING

            Ford Motor  Company
        Rouge Plant - Coke Battery
            Dearborn, Michigan
                                                    30-31
                                                    31-30

-------
                        - 258 -

                   FIGURE 8.2.1-2

A MODERATELY GREEN PUSH WITH  CONTROL SYSTEM OPERATING

                 Ford Motor Company
             Rouge Plant - Coke  Battery
                 Dearborn, Michigan
                                                         34-16
                                                          35-15

-------
                        - 259 -

             FIGURE  8.2.1-2  (continued)

A MODERATELY GREEN PUSH WITH CONTROL SYSTEM OPERATING

                 Ford Motor  Company
             Rouge Plant  - Coke  Battery
                 Dearborn, Michigan

                                                          34-17
                                                         35-16

-------
                        - 260 -
             FIGURE  8.2.1-2  (continued)

A MODERATELY GREEN PUSH WITH  CONTROL  SYSTEM OPERATING

                 Ford Motor Company
             Rouge Plant  - Coke  Battery
                 Dearborn, Michigan

                                                          34-18
                                                          35-17

-------
                   - 261 -

              FIGURE 8 .2.1-3

A GREEN PUSH WITH CONTROL  SYSTEM OPERATING

            Ford Motor  Company
        Rouge Plant.- Coke Battery
            Dearborn, Michigan
                                                    36-4
                                                   37-4

-------
                   - 262 -

        FIGURE 8.2.1-3  (continued)

A GREEN PUSH WITH CONTROL  SYSTEM  OPERATING

            Ford Motor  Company
        Rouge Plant - Coke Battery
            Dearborn, Michigan
                            36-5
           37-5

-------
                   - 263 -
        FIGURE 8.2.1-3  (continued)
A GREEN PUSH WITH CONTROL  SYSTEM  OPERATING
            Ford Motor  Company
        Rouge Plant - Coke Battery
            Dearborn, Michigan

                            36-6
           37-6

-------
                                TECHNICAL REPORT DATA
                          (Please read Inxructions on the reverse before completing}
1. REPORT NO.
EPA-600/2-77-187a
                                                       3. RECIPIENT'S ACCESSION-NO.
4. TITLE AND SUBTITLE
Emission Testing and Evaluation of Ford/Koppers
  Coke Pushing Control System;  Volume I.  Final Report
                                    5. REPORT DATE
                                     September  1977
                                    6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
          Fred Cooper,  Thomas Loch, John Mutchler,
 and Janet Vecchio  (Clayton Environmental Consultants,
 	Inc.)
                                                       8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Ford Motor Company
The American Road
Dearborn, Michigan 48121
                                    10. PROGRAM ELEMENT NO.

                                     1AB604
                                    11. CONTRACT/GRANT NO.
                                     68-02-0630
12. SPONSORING AGENCY NAME AND ADDRESS
 EPA, Office of Research and Development
 Industrial Environmental Research Laboratory
 Research Triangle Park, NC 27711
                                    13. TYPE OF REPORT AND PERIOD COVERED
                                     Final: 12/73-8/77	
                                    14. SPONSORING AGENCY CODE
                                      EPA/600/13
15. SUPPLEMENTARY NOTES IERL_RTp project officer for this report is Norman Plaks, Mail
Drop 62, 919/541-2733.
ie. ABSTRACT
              repOrj. documents a field testing and engineering evaluation of the per-
formance of a retrofitted, mobile-hood,  high-energy-scrubber control system, abating
coke-side pushing emissions from a 58-oven coke battery.  It documents the venturi-
scrubber inlet and outlet emission rates, as well as emission factors for filterable and
condensible particulate, sulfur dioxide, sulfur trioxide, total sulfur oxides, carbon
monoxide, nitrogen oxides,  benzene, benzene and homologues, total light hydrocarbons
methane and homologues, and ethylene and homologues.  Results indicate that the
mobile hood collects particulate emissions resulting from green and clean pushes at
22% to 54% efficiencies, respectively, or from 30% to 92%, respectively, using two
different estimating techniques. Scrubber performance averaged 99.3% for particulate
emissions captured in the hood.
17.
                             KEY WORDS AND DOCUMENT ANALYSIS
                 DESCRIPTORS
                                           b.lDENTIFIERS/OPEN ENDED TERMS
                                                 c.  COSATI Field/Group
Air Pollution
Scrubbers
Coke
Coking
Emission
Dust
Iron and Steel Industry
Air Pollution Control
Stationary Sources
Coke Pushing
Particulate
13B
07A
21D
13H

11G
11F
13. DISTRIBUTION STATEMENT

 Unlimited
                        19. SECURITY CLASS (ThisReport)
                        Unclassified
                         21. NO. OF PAGES

                               273
                        20. SECURITY CLASS (Thispage)
                        Unclassified
                                                 22. PRICE
EPA Form 2220-1 (9-73)

-------