United States       Office of Analysis      EPA 440/2-80-083
          Environmental Protection    and Evaluation       November 1980
          Agency         Washington, DC 20460
          Water
x>EPA     Background Document for
          Modification of pH Effluent
          Limitations Guidelines and
          Standards for Point Sources
          Required by NPDES
          to Monitor Continuously
          Effluent pH

-------
                       Contents of pH Background Document



Executive Summary and Conclusions..	 p.  I

1.   Introduction	p.  1

     1.1  Effluent Guidelines	p.  1

     1.2  National Pollutant Discharge Elimination

        System and The Continuous Monitoring Requirement	p.  2

2.   Technical Definition of pH	p.  2

3.   Factors Affecting pH Control	p.  4

     3.1  Wastewater Source	p.  6

     3.2  Pollutants Present in Wastewater	p.  7

     3.3  pH Control Systems	p.  7

        3.3.1  Number of Stages	p.  8
        3.3.2  Smoothing Capacity	p. 11
        3.3.3  Diversion Systems	p. 11
        3.3.4  Placement of pH Monitors	p. 11

     3.4  Maintenance	p. 11

4.   Historical Background	p. 13

     4.1  Legal Petition:  Description of Contents	p. 13

        4.1.1  Industrial Report Submitted to EPA	p. 14

     4.2  Agency Data Collection Activities	p. 15

        4.2.1  Initial  Agency Data Collection Activity:
               Agency Data Base I	p. 15

        4.2.1.1  Plant Selection	p. 15
        4.2.1.2  Data Acquisition	p. 16
        4.2.1.3  Date Base Description	p. 17

        4.2.2  Second Agency Data Collection Activity:
                Agency Data Base II	p. 17

        4.2.2.1  Plant Selection	p. 19
        4.2.2.2  Data Acquisition	p. 19
        4.2.2.3  Date Base Description	p. 20


                                       (i)

-------
     4.3  Treatment System Characteristics	p. 24
5.   General Approach and Rationale for Continuous Monitoring
     Compliance Criteria 	p. 25
     5.1  Introduction	p> 25
     5.2  Support of Modified pH Limitations	p. 36
     5.3  Expected Value and Variance of Total Monthly Excursion Time..p. 46
     5.4  Nonparametric Analysis of Number of Monthly Violations	p. 49
6.   Agency Response and Position on Technical Conclusions
     Submitted in 1978 Petition	p. 51
     References	p> 53
0    Acknowledgements	p. 54
0    Appendix
     -   Real/Apparent Frequency Distributions of Excursion Time by Plant
     -   Combined Agency Data Bases
     -   Cleary, Gottlieb,  Steen and Hamilton Industrial  Petition
     -   pH Under Continuous  Monitoring
                                      (11)

-------
                       EXECUTIVE SUMMARY AND CONCLUSIONS
The information presented in this report evolved in response to a petition
directed toward the Environmental Protection Agency from various industries
seeking relief from absolute compliance of the 6-9 pH categorical standard when
continuous monitoring is required.  The standard of pH 6-9 was initially intended
to apply to those situations for which discrete samples (i.e., grab or composite)
were to be monitored for pH.  This sampling procedure monitors pH at fixed
instantaneous moments in time.  As such the determination of pH values outside
the 6-9 range based on discrete sampling is dependent on when and how often the
samples are drawn.  Recently, some permittees have been required (through the
NPDES permit system) to monitor pH using continuous pH monitoring instrumentation.
This monitoring instrumentation produces a continuous 24 hour reading of pH on
a minute by minute basis.  Because of the change in monitoring techniques
values outside the 6-9 range are now recorded as well as the total duration time
outside the range and the maximum/minimum pH values outside the 6-9 range.
According to the petition the standard of pH 6-9 intended for discrete sampling
situations requires modification because of the introduction of this more
thorough monitoring schedule.  Two Agency data collection activities in conjunc-
tion with Agency and industry data analyses form the basis for the modified pH
regulation.  The Agency contends that well-operating and properly maintained
plants when required to monitor pH continuously, can maintain compliance with
the categorical pH standard 6-9 at least 99 percent of the time on a monthly
basis and that individual excursions should be limited to no more than 30 minutes.
However, local water quality conditions may require setting an effluent pH
limitation considerably less than the thirty minutes proposed.  Permit writers
may impose more stringent limitations when local water quality conditions
warrant such action.  The Agency evaluation and response to the petition is in
Section 6.

In Section 1 of this document the absolute pH regulation is discussed with
respect to discrete and continuous monitoring situations.

Section 2 of this document presents a technical definition of pH.

In Section 3, factors which affect pH control are  discussed.  The section
addresses the factors of wastewater sources, pollutants, pollution control
systems, and maintenance in relation to pH control.

In Section 4, the major activities which resulted  in the Agency's development
of the modified pH standard are  presented.

Section 5 presents the results from examining the  two Agency data bases with
respect to long term compliance  and monthly compliance for the modified pH
compliance specifications.  Additionally a plant specific description of the
frequency of events outside the  6-9 range  (identified as excursions) exceeding
the proposed Agency individual excursion duration  time limit  is  presented.

Section 6 presents the Agency response and position on the technical
conclusions submitted in the 1978 petition.

-------
                               1  INTRODUCTION

                            1.1  Effluent Guidelines
In 1972, the Federal Water Pollution Control Act was amended to establish a
regulatory system with abatement requirements, enforcement procedures, and
penalties for violations.  The 1972 amendments required the Environmental
Protection Agency (EPA) to establish national effluent guidelines for both
municipal and industrial  dischargers.

Effluent guidelines which go.vern levels of pollutants in discharges of indus-
trial wastewaters developed under these amendments for the Best Practicable
Control Technology (BPT)  affecting pH were first proposed in 1974; for almost
all of the guidelines, the standard was to maintain control of pH at levels
between 6 and 9 of discharged wastewaters.  Examples of exceptions to the 6-9
standard are; secondary aluminum smelting of non-ferrous metal manufacturing
which is required to maintain its effluent pH within the range 7.5 to 9 and
deflourinated phosphate rock of the phosphate manufacturing category is required
to maintain its effluent pH between 6 and 9.5.  However for purposes of this
document effluent pH within the 6-9 range is the categorical standard of concern,

To comply with these guidelines, certain industrial plants including inorganic
chemical manufacturing, iron and steel manufacturing, and fertilizer manu-
facturing—installed equipment designed to control pH at this level.  Initially,
to set a basis for BPT, data on pH were obtained via the use of either grab
samples or composite samples.

Grab samples, taken at a given instant in time, represent the conditions
that exist at the moment of sampling and not necessarily the conditions at any
other time.  These samples are often used to corroborate the results of compo-
site samples.  A composite sample is made up of a number of discrete samples
taken during a predetermined time period.  It therefore consists of a series
of grab samples.  The discrete samples comprising the composite sample are
collected at equal time intervals.  If the flow rate of the effluent is con-
stant, equal volumes of each discrete sample are combined.  If flow is variable
then these volumes are in proportion to the flow rate.  Composite samples are
used to characterize highly variable waste streams and to provide average
discharge concentrations.  In this document all pH values lower than 6 or
higher than 9 are defined as excursions.  When grab or composite samples are
used, the detection of a pH excursion depends upon the samples being drawn
simultaneously with the occurrence of the excursions.  Even if an excursion
were detected, there would be neither assurance that the sample was obtained
at the maximum point nor data to determine the duration of the excursion.

-------
              1.2   National  Pollutant Discharge Elimination  System
                    And Continuous Monitoring Requirements


The National  Pollutant Discharge Elimination System  (NPDES), is the  vehicle
through which the  national  effluent guidelines are implemented at industrial
plants.  The  system requires that industries discharging  into navigable waters
obtain NPDES  permits which  limit the kinds and quantities of pollutants that
may be discharged.   These  permits are issued at the local  level, based on the
national standards, and take into consideration for each  individual  plant
economic factors,  technical feasibility, effluent characteristics, quality of
the receiving water, the use classification of the receiving water,  and state
water quality standards.

When effluent pH is highly  variable, an NPDES permit may require that continuous
pH monitors be installed.   A continuous monitor is essentially an electronic
probe, located in  the effluent, connected to an amplification unit and a recorder.
The recorder uses  a moving  chart and a pen which responds to the amplified
voltage signal, continuously tracing the pH level on the chart.  There are two
advantages of using continuous monitors.  First, continuous monitoring ensures
the observation of all excursions outside the pH 6 to 9 range;  and second, the
duration and peak  value of  each excursion are measured.  Even for an excursion
which is of short  duration, a narrow, sharp peak is recorded.  The same peak
would probably be missed if grab samples were used.

The results of continuous pH monitoring differ from those obtained by discrete
sampling because of the monitoring procedures.   Discrete samples are individual
values of wastewater taken  for the purpose of chemical  analysis and  reporting.
These discrete samples may  be taken  by grab or composite method.   Grab samples
are drawn at a single instant in time.   A composite sample consists  of a series
of three or more smaller samples taken  over a specified time period  and combined
into a single, larger sample for a single analysis or measurement.   Because
they are not taken continuously, discrete samples (both grab and composite)
do not measure all excursions.   Moreover, composite samples may mask short-term
excursions by blending a single excursion sample with others which do not
indicate excursions.   Breakdowns and errors on  the part of process plant and
treatment system equipment can  cause discharge  of effluent outside the permissible
pH range.  Continuous monitors  will  record all  excursions resulting from these
and other conditions.

Problems with pH measurements obtained  by continuous monitoring results
stem from either monitor breakdowns  or  instrument and calibration  errors.   The
former produces no results, while the latter produces spurious  results, includ-
ing apparent excursions when none occurred or apparent  compliance  when pH was
outside the limits of 6 and 9.   Apparent excursions will  be discussed later in
this document.

                      2.   TECHNICAL  DEFINITION  OF pH


pH is  a measurement of the concentration  of an  acid or  base present,  in an
aqueous solution,  expressed as:

-------
          pH = -logio(hydrogen ion concentration in moles/liter)

     or

                                 pH = -log10(H+)

Autoprotolysis of water results in a hydrogen (or hydronium) ion concentration
and a hydroxyl ion concentration, (OH~).  Using the same approach, the
concentration of hydroxyl ion can be expressed by the indicator pOH as follows:
                               pOH = -log10(OH-)
For water, the product of the
concentration is a constant
of these terms is:
  he hydrogen ion concentration and the hydroxyl ion
  , Kw, which is equal to 1.0 x 10~14.  The relationship


Kw = (H+) (OFT) = 1.0 x ID'14
and therefore

                       PKW = -logioKw = pH + pOH = 14.00
A neutral solution contains an equal concentration of (H+) and (OH") ions and
thus satisfies the following conditions:

                           (H+) = (OH-) = 1.0 x 10-7

                                pH = pOH = 7.00

pH values of less than 7 are acidic; they have a higher hydrogen ion concentration
than a neutral solution, while pH values greater than 7 are basic.  In water,
if pOH equals zero, the pH equals 14, thus maintaining pKw at 14i/.  Because of
this property of water, pH measuring devices used for wastewater monitoring
have a range of pH = 0 to pH = 14.

pH monitors do not measure the hydrogen ion concentration directly.  Instead,
they measure the hydrogen ion activity.  This relationship is expressed as:!/

                             E = A + B logic UH+)

                        where E is the pH meter voltage,

                           A and B are constants, and

                            is the hydrogen ion activity
However, calibration of pH meters using known concentrations of hydrogen ions
provide an accurate record of the hydrogen ion activity that is useful in
determing both the degree of wastewater neutralization and the need,  if any,
for additional treatment.

-------
                   3.   FACTORS AFFECTING  pH  CONTROL


The factors that  can influence both the need for  pH control  and  the  performance
of pH  controllers are:

     0  Sources of plant wastewater
     0  Pollutants present  in wastewater
     0  Type of pH control  system
     0  Maintenance of  pH control system.

It is  important to recognize the principles  that  determine control of  pH  and
to review the technical factors that affect  pH control.   In  general, pH of  a
wastewater stream is controlled by adding acid to a basic waste  stream or base
to an acidic waste stream.  Exhibit 1 contains typical neutralization  curves
that illustrate the principles of pH control.  These curves  compare the amount
of acid added (to neutralize a basic solution) with the corresponding  decrease
in pH.  A similar figure can be drawn to demonstrate neutralization of an
acidic solution by adding base.  The three curves illustrate the types of
situation requiring pH control which occur in industry.  The sodium hydroxide
(NaOH) curve represents the situation which  is most difficult to control.
The curve has three zones:   Zones A and C showing little or no change  in pH
with the addition of acid, and Zone B showing an extreme change  in pH with the
addition of acid.  The 6 to 9 range is included in the extreme-change area,
Zone B.  This type of situation is difficult to control, since the response
to added acid is either extreme (steep slope) or nearly nil  (low slope).  The
ideal  situation for control  is represented by a linear medium-sloped response
to acid addition.

-------
I'll    12-
     10-
                                                                                                                          X
                                                                                                                          PC
                                                                                                                          M
     2-
     0-
                                              Increase in Amount of HCl Titrant
                                  H
                                  H-
                                  rt
                                                                                                                          P-
                                                                                                                          O
                                                                                                                          C"

                                                                                                                          §
                                                                                                                         OJ
                                                                                                                          o
                                                                                                                          O
                                                                                                                         U)

                                                                                                                          tu
                                                                                                                          P
                                                                                                                          o-
                                                                                                                          O
                                                                                                                          O
                                                                                                                         U)
                                                                                                                          o
Relative Volume  (Units  depend upon the concentration  of  HCl, NaOH,
NH
                                                                                          o» and container volume)

-------
The ammonia  (NH3) curve represents a less extreme, therefore less difficult,
control situation than the sodium hydroxide curve.  The overall range of the
ammonia curve is narrower than the range of the sodium hydroxide curve; Zone A
of the ammonia curve is steeper than Zone A of the sodium hydroxide curve; and
the transition from Zone A to Zone B of the ammonia curve is within the 6 to 9
pH range.  These three factors indicate that with the addition of small amounts
of acid, in Zone A, the pH level of ammonia will change more than the pH level
of sodium hydroxide, but in Zone B, the pH of ammonia will change less than
the pH of sodium hydroxide.  Nevertheless, these pH changes in the 6 to 9
range of both solutions are significant, and control can be difficult to achieve.

Of the three situations illustrated, the sodium carbonate curve represents
that which is easiest to control, as indicated by its moderate slope in all
zones.  In this case, the addition of small amounts of acid causes moderate
changes in pH in all zones, at all pH values, including the 6 to 9 range.

Both ammonia and sodium carbonate act as buffers; the phenomenon of buffering
can have a major effect on pH control.I/  A buffered solution tends to resist
a change in pH.  The presence of buffers can make treatment of effluents easier,
depending on the pH of the incoming wastes and the buffering range of the
pollutants.  A solution with buffering capacity, such as one containing sodium
carbonate (Na^COs), has an inherent resistance to changes in pH.  Therefore,
a large quantity of acid would be required to effect a change in the pH of a
Na2C03 solution.  On the other hand, a much smaller quantity of acid would
be required to cause a change in a sodium hydroxide (NaOH) solution.  A pH
response curve reflecting this phenomenon would be a moderate-slope, nearly
linear curve, rather than an extreme S-type curve.  Because of its tendency to
moderate pH extremes, wastewater with buffering capacity permits easier pH
control than does wastewater without buffering capacity.

                            3.1  Wastewater Sources


Wastewater sources can influence the performance of a pH control system.
These systems rely on mechanical devices to control effluent pH.  Because
mechanical  devices cannot respond quickly enough to severe fluctuations in
wastewater acid or base loadings, systems which employ these devices are
correspondingly limited in their ability to control pH levels.!/!/  Sources of
wastewater which have acid or base loadings that are highly variable can
create problems in pH control.  Often, installation of large equalization
volumes!/ is necessary to balance the acid loadings.  Equalization volume
consists of a tank or basin used to accumulate wastewater flows from two or more
sources.  In this step, fluctuating flows are combined in order to balance or
equalize the final pH level.  Sources of wastewater with relatively constant
acid or base loadings are generally easier to control  by the use of mechanical
systems.!/!/

Spills and intermittent dumps of acids or bases are common examples of industrial
sources of abrupt changes in wastewater pollutant loadings.   Ion exchange
regeneration, consisting of batch discharges of strong acid and base, is another
common example of variable waste loading from industrial sources.   A strong-acid
leak into non-contact cooling water is a more dramatic example of variable

-------
loading.  When a leak occurs, the non-contact cooling wastewater stream may
experience surges of low (or high) pH which can exceed the neutralizing capacity
of the pH control plant.  These surges are know as "shock loads."!/ In some
plants, rainwater runoff can be another source of wastewater producing variable
loading.  There are several examples of industrial sources of nearly constant
acid or base loadings.  These constant load sources are easier to control, as
the systems need not respond to highly variable load changes.  Caustic degreasing
overflow is an example of an industrial constant-load wastewater source.
Spent acid or caustic treating solution is a fairly common wastewater source
in petroleum and organic chemical operations.  Similarly, spent acid or caustic
scrubber overflow is another example of a constant-load wastewater source in
several industries.  Wastewater sources such as the three described above are
less difficult to control than those with highly  variable effluent.  Simple
systems for their control can be designed which do not require the capacity to
respond to wide fluctuations in load..!/!/

Previous discussions of sources of wastewater demonstrate the strong influence
these sources exert on the effectiveness of pH control.  Many plants have both
constant and fluctuating acid or base  loadings.   The variability of the aggre-
gate load would therefore determine the control system design.  Plants having
relatively constant loading  in the influent wastewater can utilize fairly
simple treatment systems.  Plants having extremely variable  wastewater loading
must employ equalization volume and possibly extra equipment to provide the
required flexibility to assure adequate pH control.


                     3.2   Pollutants Present in Wastewater


Pollutants present  in  influent wastewater  can have a substantial effect on  pH
control  system  performance.  Many pollutant compounds  are buffers  and can make.
control  easier  to achieve  (Exhibit 1).  Pollutants can be either acidic buffers
or  basic buffers.   Both types of buffer resist pH changes.   The addition  of
extra neutralizing  reagent  is required whenever buffers  are  present in the
wastewater because  buffers  resist pH changes.  This  same resistance, however,
tends  to lessen  the response of the effluent pH to the effects of  neutralization.
Ammonia, carbonate, and acetate  ions are  examples of basic buffering pollutants.
Ferric,  nickel,  bisulfite,  and phosphate  ions are examples of acid buffers.
Zinc  and aluminum  ions  serve as  both acidic and basic  buffers.


                             3.3   pH Control Systems


The design and  operation  of  a  pollution control system has considerable  impact
on  its  performance.  Several authorsl/A/  have written  about  systems to  control
pH  in  wastewater by neutralization.  They emphasize  the  significance  of  control
mode  (feed-forward  or  feedback),  number of stages, and size  of  smoothing  capacity.
An  earlier study!/  suggests  additional considerations:   the  placement  of  a
pH  monitor,  and the presence or  absence of a  diversion system.  The following
system design  parameters  are important:

-------
     0  Number of stages
     0  Smoothing capacity
     0  Diversion systems
     0  Placement of pH monitors.

3.3.1  Number of Stages

The two standard pH control references by Moorel/ and Shinskeyl/ recommend
mulistage neutralization over single-stage neutralization for pH control.  In
single-stage neutralization (Exhibit 2), if too much or too little reagent is
added to the neutralization volume, fluctuations are passed directly into the
effluent.  In this case, the neutralization volume must be made extremely large
to reduce these fluctuations, especially in an unbuffered solution such as
the one containing sodium hydroxide.  Moorel/ claims that placing a second
stage of neutralization after the first stage (Exhibit 3) offers several advant-
ages.  The second stage reduces fluctuations occurring in the first stage at
the "knee" of the pH response curve (Exhibit 1), allowing the effluent pH a
greater degree of sensitivity to the addition of neutralization reagent.
Although operation at a "knee" also causes more fluctuations, these fluctuations
are reduced in the second stage.  Therefore, superior coarse pH control occurs
in the first stage, and fine or trim control is achieved in the second stage.
Even finer control can be achieved by the addition of a third stage.  The
Agency's 1979 study!/ found that those plants having multi-stage neutralization
tended to achieve a pH range of 6 to 9 more consistently than those having
one-stage neutralization.

Moorel/ points out one disadvantage of mulitstage pH control:   The second-
stage vessel  should be larger than the first-stage vessel to avoid amplification
of fluctuations.   Moorel/ recommends a 5-to-l ratio of second-stage to first-
stage volumes.  This ratio may be made somewhat smaller if sufficient buffering
capacity is present in the wastewater.  Nonetheless, construction and opera-
tion of large tanks or basins of this sort can be costly.!/!/

-------
EXHIBIT 2:  SINGLE-STAGE NEUTRALIZATION SYSTEM
                      .Valve
Caustic _


stment ^^^
Valve
tment ^^
Valve


i j-



\
r
^
\






'p\
•0
i
- i




\
t







j£
"'




C
r^
LL







}
4*
1J
M^
,pn neter
! /

\
— • \ . '. f r f -I _ *.

\
pH Electrode


             Mixer'

-------
   EXHIBIT  3:  MULTISTAGE NEUTRALIZATION SYSTEM
Caustic

   Acid _
   Valves
                     pH Controllers
Stage 1
        Mixers
                                        Stage  2
                                                   pH Electrode

                                                     Effluent
                        1U

-------
3.3.2  Smoothing Capacity

Moore!/ and Shinskey!/ also recommend use of a smoothing tank or basin as a
final step before release of the treated effleunt.  As the term implies, the
smoothing stage tends to damp out or smooth excursions in effluent pH. Again,
the presence of buffering chemicals in the wastewater can reduce the need for
smoothing capacity, as the buffers themselves damp pH fluctuations.  Several
plants included in the 1980 EPA study!/ used smoothing basins before releasing
their final effluents.

3.3.3  Diversion Systems

Diversion systems are "systems of last resort" which are used whenever a treat-
ment system cannot treat the wastes fully.  Diversion systems reroute effluent
flow which is outside the 6 to 9 pH range into a  diversion tank or basin which
holds the effluent until it can be recycled through the treatment system.
Moore!/ points out that diversion systems are of  little value unless the
holding capacity of the basin is sufficient to allow time to correct the problem
that is affecting the effluent pH.  Several plants visited during the 1979  EPA
study!/ employed diversion systems.  These  systems were designed to  accumulate
effluent flow for several hours.  The  length of  holding time was determined to
allow plant personnel to locate and repair  the unit causing the pH control
problem.

3.3.4  Placement of  pH Monitors

In  plants  using diversion systems,  placement of  pH monitors has an  important
effect on  recorded results.   If a plant's pH monitor were  located  where  the
effluent would  gather when flow is  diverted, the pH recorder would  continue to
indicate an effluent  pH  reading,  even  though no  effluent was  being  discharged.2/,
These  apparent  excursions are defined  to  be "technical  excursions."  Several
plants visited  during the 1979 EPA  studies!/!/  had  recorded technical excursions,
The recording  of  such excursions  can  be  avoided  by  the  proper  location  of  pH
monitors.


                                3.4  Maintenance


Maintenance  is  required  for  all  components  of  a  pH  control  system.   Lack of
 adequate maintenance can result  in  erratic  or  poor  pH  control.

Continuous control  of effluent  pH is  achieved  by the  use of a "pH  recorder-
 controller."   This  system consists  of three parts:   (1) pH electrodes,  (2)  a pH
 recorder,  and  (3)  a  pH  controller (Exhibit  4).   Electrodes produce an  electrical
 signal  (voltage)  that is sent to  the  recorder  and controller.   This voltage
                                        11

-------
            EXHIBIT  4:  pH  RECORDER-CONTROLLER
[    Reagent
      1
  PH
/Electrodes
                                -»*{ Controller J
                                  >• [  Recorder
   Waste Stream
                                          »pH Chart
                             12

-------
is proportional to the pH of the water In which the electrodes are placed.
The recorder contains a pen and a chart drive.  The position of the pen changes
in relation to the magnitude and sign (+ or -) of the pH signal.  The chart
drive moves the paper at a specific rate.  The controller contains a comparision
unit which compares the magnitude and sign of the incoming pH signal to the
value of a single preset control point (set point).  The controller then  produces
another electrical signal proportional to the difference between the incoming
pH signal and the set point.  This controller signal is sent to the reagent
unit arid determines the amount of neutralizing reagent that is added.

Maintenance of electrodes is particularly important to both the precision and
accuracy of pH control and monitoring.  Three types of electrode maintenance
are required:  (1) calibration,  (2) cleaning, and (3) replacement.  The voltage
output of pH electrodes tends to drift when electrodes are left in continuous
service.  Calibration compensates for this drift.  Calibration is accomplished
by immersing electrodes  in a known pH solution and adjusting the electrical
signal to give a  reading corresponding to the solution's pH.  Periodic calibra-
tion is recommended,  and some plants  calibrate as often as once a day.  Solids,
tars, and oils present in waste  streams can hinder the performance of electrodes
and adversely  affect  pH measurement.  Electrodes exposed to these substances
must be cleaned  periodically to  maintain their accuracy.  The necessary fre-
quency of cleaning depends upon  the nature of the waste stream.  When  in  con-
tinuous use,  pH  electrodes deteriorate over time; therefore, calibration  is
required more  frequently as tine in  service  increases.  When electrodes are
replaced periodically, the need  for calibration  is  less frequent.

Recorders  and  electrical signal  systems  require  periodic  voltage  and  current
checks;  recorder pens and  chart  drives must  be inspected  and oiled  regularly;
and unless  they  receive  frequent maintenance, the  performance  of  controllers
will  be  affected adversely.  Maintenance,  therefore,  is an  important  factor
affecting  pH  control  system performance.   Proper periodic  calibration,
inspection,  cleaning, and  replacement of  critical  system  components  have  a
significant  impact  on both  the  recording  and  control  performances of  a  pH
control  system.   Lack of  adequate  maintenance can  result  in  inaccurate
recordings  and unreliable  control.

EPA recommends that  continuous  monitoring  be  serviced  and  calibrated  as  per
the manufacturers specifications or  guidance  in  conjunction  with  considera-
tion  of  the characteristics of  the wastestream  being  monitored.


                            4.   HISTORICAL  BACKGROUND

                  4.1  Legal  Petition:  Description  of Contents


 In August, 1978, the Environmental  Protection Agency  received  a petition
from Cleary, Gottlieb, Steen  and Hamilton regarding "pH Effluent  Limitation
Guidelines and Standards of Performance for Certain Manufacturing Point Source
 Categories."  This petition  requested the revision of certain  EPA pH effluent
 limitations guidelines for existing plants and  of standards of performance for
 new sources promulgated under authority of Sections 301(b), 304(b)  and 306(b)


                                        13

-------
 of the Federal  Water Pollution  Control  Act.   Specifically,  the Agency was
 requested to (1)  rescind the Agency's  internal  policy limiting the pH of effluent
 discharge to a  range of 6.0 to  9.0  on  a continuous  basis  for industrial  point
 source categories,  and (2)  revise the  pH parameters in  effluent limitations
 guidelines for  existing plants  and  revise the standards of  performance for new
 sources to permit excursions outside the 6 to 9 pH  range  for a total  period of
 at least one percent of a month, with  excursions below  pH 3.5 or above pH 11
 being  limited to  15 minutes per excursion.

 The categories  and  subcateggries listed in the  petition included (but were not
 limited to )  the  following:!/

     1.   Inorganic  Chemicals (40 CRF Part 415):  Subpart D
          (Calcium Chloride); Subpart F  (Chlorine and  Sodium
          or  Potassium  Hydroxide); Subpart G  (Hydrochloric
          Acid); Subpart H (Hydrofluoric Acid):  Subpart  J
          (Nitric  Acid);  Subpart 0 (Sodium Carbonate); Subpart U
          (Sulfuric  Acid): Subpart V (Titanium Dioxide):
          Subpart  W  (Aluminum Fluoride Production);  Subpart Y
          (Ammonium  Hydroxide),  Subpart  AP (Hydrogen Cyanide);
          Subpart  AV  (Strong Nitric Acid);  Subpart BF  (Sodium
          Silicofluoride  Production)

     2.   Fertilizer  Manufacturing (40 CFR Part  418):  Subpart  B
          (Ammonia);  Subpart  C (Urea);  Subpart D  (Ammonium
          Nitrate);  Subpart  E (Nitric Acid)

     3.   Iron and Steel  Manufacturing (40  CFR Part 420)

     4.   Non-Ferrous Metals  Manufacturing  (40 CFR Part 421):
          Subpart A  (Bauxite  Refining);   Subpart B  (Primary
          Aluminum Smelting); Subpart C  (Secondary Aluminum
          Smelting)

     5.   Phosphate Manufacturing (40 CRF  Part 422):   Subpart A
          (Phosphorus Production);  Subpart B (Phosphorus
          Consuming)

     6.   Organic Chemical Manufacturing (40 CFR Part 414):
          Subparts covering  plants manufacturing, or using,
          in manufacturing process,  strong acids or bases.

4.1.1  Industrial  Report Submitted  to  EPA

In conjuction with the petition, deary et al., submitted  a  report on the
performance of pH systems at several  industrial facilities.^/  The report
concluded:

        Control  of pH in the 6 to  9  range on  a continuous  basis is extremely
        difficult for highly acidic  raw wastes
                                       14

-------
        A compliance level  of 99 percent achievement of pH 6 to 9 is reasonable
        on a monthly basis;  excursions outside the pH 3.5 to 11 range can be
        limited to less than 15 minutes

        Number of excursions is not a valid measure of pH control system
        performance
        Costs rise rapidly as plants attempt to improve performance to level
        exceeding 99 percent (of time) per month.
s
The industry report did not, however, contain a uniform data base or analysis
to support conclusions regarding plant performance under continuous PH monitoring.
The Agency therefore undertook a study to assess the validity of the conclusions
submitted by industry.  Further, this study would document both the performance
of the control systems and their recorded capital and operating costs.


                     4.2  Agency Data Collection Activities


In order  to examine and evaluate control systems currently in use, the Agency
conducted the two  data collection  and analysis  activities described below.

4.2.1   Initial Agency Data  Collection Activity:  Agency Data Base  I

The Agency  collected  and  analyzed  data  to  aid its  assessment of  current  pH
control  systems.   A pH data base was constructed from  six plants  in various
categories  and subcategories in  several  industries.  The  pH  data  was  used to
generate plots,  cross  tabulations  and tables  which  illustrated  relationships
among  the data in  support assessments of pH control  for the  various plants.

4.2.1.1   Plant Selection.  The selection of plants was a  primary consideration
 in the development of this  data  base.   Sixty-one candidate  plants were identified
from  a 1978 report^/  and  by EPA regional offices and State  water enforcement
 personnel.   Nine plants  were selected,  based upon  the  following criteria:

      0  Treatment  systems which aim primarily at controlling pH
      0  No biotreatment
      0  Inclusion  in  raw waste stream of process water or contact cooling water
      0  Recommendation of treatment system as well-operating
      0  Distribution  among several  industries.

 An accurate assessment was ensured  by examining treatment systems that
 control only pH.  Treatment systems for other effluent characteristics were not
 considered.  The technology employed by plants considered in this study  is
 single and multistage neutralization and ion exchange.  In the ion exchange
 process, an acid or base in the form of an insoluble resin is used to treat
 wastewater.  This process contrasts with conventional neutralization, which uses
 soluble  acid or base for treatment.  Since both processes effectively achieve
 either acid or base neutralization  of the wastewater  solution, the results  of
 effluent treatment by ion  exchange  are  similar to those of treatment by
 standard neutralization.


                                         15

-------
 Plants  utilizing  biotreatment  as  a  primary treatment  method were not included
 in  this  analysis  because  the technology  is fundamentally different  from that
 of  standard  neutralization  and because pH  control  is  often  incidental  to the
 purposes of  biotreatment.   Wastes subject  to  biotreatment are  neutralized
 before  treatment.   This neutralization does not  need  fine control,  since bio-
 treatment  itself  causes pH  changes.   These pH changes are variable  and depend
 both  upon  the  nature  of the plant wastes and  upon  the type  of  biotreatment tech-
 nology.

 Only  plants  operating  at  an acceptable level  of  effectiveness  were  selected to
 be  studied,  since the  results  of this study were intended to develop standards
 which all  pH control  systems can achieve through adequate design, operation, and
 maintenance.   Plants  on the candidate list were  contacted by telephone for
 further  screening,  using  the following criteria:

         Operation of  an automatic continuous  pH  recorder to  monitor
         treatment of  effluent  streams

         Availability of at  least 6 months  of  pH  recording charts  for treated
         effluent

      0   Operation of control system during the period  of  a  plant  visit;
         cooperation with  such  a visit.

 Since continuous monitoring of  pH was the  subject  of  this study,  only  plants
 with  this  type of monitor were  selected.   Since  a  primary focus  of  this  effort
 was examining  pH control performance over  a long term, the minimum  period  of
 data  desired was 6 months.  Observation of  actual  plant  operation was  necessary
 in  order to  appreciate the  performance of  pH  control  equipment.   Seven  plants
 were  selected  for visits, based upon the criteria  listed  above.  All seven
 plants were  visited between November 28, 1978 and December 18, 1978.   However,
 pH data could not be obtained from one of  these  plants because the  State of
 California (in which the plant  is located)   does not require  a plant  to  retain
 strip charts once a violation is reported.

4.2.1.2  Data Acquisition.  At each plant  investigators extracted data  from  pH
 recording charts.  Excursions above and below the pH range 6 to 9 were entered
 on the data forms.

     Data recorded during the plant visit  phase of this  study were:

     0  Date of excursion (month, day, year)
     0  Time (on a 24-hour clock) of excursion
     0  Maximum/minimum pH of excursion
     0  Duration of excursion
     0  Reason  for excursion.

For each excursion a height equal  to the peak  height of the  excursion curve
and a width equal  to the base of the excursion curve were recorded.   An example
of this extraction procedure is shown in Exhibit 5.  This method provided a
rapid, consistent means of extracting data.  An excursion was recorded on data
forms whenever the trace on a continuous pH recording chart   indicated a pH


                                       16

-------
greater than 9 or less than 6.  All types of excursions were recorded, regardless
of cause, including those resulting from instrument error and calibration.
Details can be found in the referenced report.!/

4.2.1.3  Data Base Description.  The raw data from each observed excursion were
compiled into machine-readable form.  Month, day and year were entered as
separate fields for each excursion.  Time was entered, using 24-hour clock
notations.  The maximum/minimum pH of each recorded excursion were entered, as
well as the duration of the excursion in minutes.  Reason codes were assigned,
based upon reasons recorded on the data sheets for each excursion, according
to the following coding system:

     Reason Code I                            Reason Type

         1                                    Process Upset

         2                                    Treatment System Upset

         3                                    Technical Excursion—Excursion
                                              recorded, but no flow to sur-
                                              face water was recorded.

         4                                    Other

         5                                    Unknown

         6                                    Instrument Error

 In  this  coding  scheme,  process  upsets are  unexpected  changes in the manufacturing
 process  that  can cause  a  change  in  the  treatment  system effluent  pH.  An  acid
 leak  is  an  example  of  a process  upset.   A  treatment  system  upset  is a malfunction
 of  the treatment  system that  causes  a change  in  the  effluent pH—for  example,
 treatment  chemical  pump failure.   A technical excursion occcurs when  the  recording
 chart  indicates a  pH  excursion,but  other records  indicate that  there  has  been
 no  effluent flow from the  plant  during  this  period.   This type  of  excursion
 was observed  only  at  plants  with diversion systems,  where it occurred when the
 diversion  valve was in  operation.   Other excursions  are those which occur for
 reasons  that  are known, but  are  not  included  in  Reasons 1,  2, 3,  or 6—for
 example, an excursion  caused  by  treatment  plant  shut dov/n because  of  flooding
 conditions.   The unknown  reason  code was  assigned to excursions for which a
 reason was neither recorded  nor obtainable.   When an excursion  is  indicated  on
 the recording chart because  of pH monitor  malfunction or  calibration, the
 reason  code designating that recording  is  instrument error.

 Reasons  3  and 6 of Reason  Code I represent situations in  which  an  excursion
 is  indicated on the recorder chart,  but an excursion may  not  have  actually
 occurred.   These observations can be considered  "non-real"  or  "apparent." The
 remaining reasons  under Reason Code I (1,  2,  4,  and 5)  are  considered "real"
 because  they represent observations made of an  actual excursion.

 4.2.2  Second Agency Data Collection Activity:   Agency Data Base  II
                                        17

-------
                                 EXHIBIT 5:




Illustration of Data Extraction Methodology Used for Obtaining Data Base I



/t

/
J

1
L
Peak Level pK

                               J
                                                       (=in.)
                                                                    pK - 9
                                                                    pH - 6

-------
4.2.2.1  Plant Selection.  EPA commissioned a second study, the results of which
were presented by EPA in October, 1979J>/  The objectives of the study were
to describe pH treatment systems in the inorganic chemicals industry, to estab-
lish a data base on pH compliance time within the present 6 to 9 limits, and
to perform analyses on that data base.  This study was performed to present an
assessment of pH control in the inorganic chemicals industry.

To achieve these objectives, nine subcategories for this industry were initially
chosen because the raw wastes from their production processes would likely
present pH control problems.  However, because of other considerations, some
potentially appropriate subcategories could not be included.

During the course of the study, sodium dichromate was dropped from the list of
nine subcategories, while chlor-alkali was added.  The final list of  subcategories
studied is as follows:

                    Aluminum Fluoride
                    Chi or-Alkali
                    Hydrofluoric Acid
                    Hydrochloric Acid
                    Hydrogen Cyanide
                    Sodium Bisulfite
                    Sodium Silicate
                    Sulfuric Acid
                    Titanium Dioxide  (Chloride Process)

     Plant visits were  arranged on the basis  of the following criteria:

     0  Plant possession of automatic recorder(s) that continuously monitor
        all final effluent streams at each  point  of discharge

     0  Availability  of  at least  six  months', but preferably one year's  pH
        recordings for  discharged  effluents

     0  Plant possession of excursion records

     0   In some  subcategories,  the type  of  process  used  in  production was  also
        a  factor in  selection.

 Because of process requirements  and  of various  state  or  local regulations,
 the  number of plants  in  any given  subcategory meeting the  preceding  requirements
 was  often  limited; and  in  the  case of sodium dichromate, no plant  could  be
 found  that met  the criteria.

 After  preliminary telephone contact,  plant  selections were made;  visits
 were scheduled;  and  the process  of data  collection  was  begun.

 4.2.2.2   Data Acquisition.  Three objectives were set for  each  of  the plant
 visits.   The  first and  primary objective was to  tabulate  excursion  data, includ-
 ing  explanations.  The  second  was to review the  wastewater treatment systems
 as they  affect  pH and the  raw  waste  characteristics  as  they relate to pH in  an
 attempt  to  evaluate  data from  the actual  systems  involved.   The final objective,


                                        19

-------
secondary in nature, was to obtain information on other factors affecting pH
control, such as the pH and volume of other process wastes being treated, type
of pH equipment used, and costs involved.

As in the earlier Agency study, the data were recorded in tabular form to
include date, time, maximum (minimum) pH values, duration, reason, and remarks
for each pH excursion.  The data for each plant were later revised as necessary
and analyzed with respect to the various types of excursion.

The peak value of the pH and the duration of the excursion are the pertinent
factors for recording each excursion.  The method used for extracting these
values is similar to that used in the Agency's first study.  For each excursion,
the maximum (minimum) pH was recorded as the pH for the entire duration, regard-
less of multiple peaks, as long as the pH recording line never reentered the 6
to 9 bounds.  This process is illustrated in Exhibit 6A.  The duration was
measured from point of leaving the control  range to point of reentering the
control range.  If, as a result of instrument error, the recording oscillated
above and below the bounds for a short period of time, measurements were
necessarily taken in a different manner (Exhibit 6B).  In this case, time
above pH 9 and below pH 6 would be divided into two excursions.  Peak values
are considered to be the minimum and maximum pH values observed during the
excursion.  This treatment of instrument error differs from that used in compiling
Data BASE I.  In DATA BASE I,  excursions such as the one shown in Exhibit 6B
would be recorded as a series of separate peaks (11 above pH and 12 below pH 6).

4.2.2.3  Data Base Description.  Raw data were collected, coded, and processed
in a manner similar to that used in the Data Base I study.  Since these two
data bases were compiled in separate and distinct activities within the Agency,
the reason codes used are not identical.  The reasons for excursions were
coded in Data Base II as follows:
                                       20

-------
Reason Code II                   Reason Type

     1                           Process Upset

     2                           Treatment System Malfunction/Shutdown

     3                           Instrument Error

     4                           Instrument Calibration/Maintenance

     5                           Operator Error

     6                           Diversion in Operation,  but pH Monitor
                                 Still  Recording (flow stopped or diverted,
                                 but the position of the  pH probe caused
                                 recording pH of water that was not being
                                 discharged)

     7                           Other  (apparent only)

     8                           Unknown

     9                           Emergency Operation

    10                           Spills or Leaks

    11                           Rainwater Overflow

    12                           Other  (actual)
                                 21

-------
  EXHIBIT 6:  Illustration of Typical Excursions and Data
              Extraction Methodology Used  in Obtaining Data
              Base II


                 A.   Tvoical  Excursion Peak
                                          Instrument Check
                                            H 11
                                           FH9
                                           tin 6
                                                  Date 1-1-79
               12:00     12:15      12:30
                         Time
The example excursion above would have beer, recorded as:
                                             McriTj"1 — JajTuarv 19 7 9
DP-JTE      TIME     PEAK    DURATION (ir.ir.)        PZASON
1-1       12:00    11.0          30                4
                 B.   Non-Steadv Excursion
                                             K 10

                                             H 6
                    12 Minutes
A maximuiE "total"  excursion of  six minutes duration with  a  peat:  pH
of 10 along with a ciniaua "total" excursion of six minutes with
peak pH of 5 would have  been recorded.
                               22

-------
 These reason codes are defined as follows:

 1.  Process Upset - production problems or unexpected interruptions in
     production resulted in pH excursions.

 2.  Treatment System Malfunction/Shutdown - failure of the treatment system
     itself to handle wastes properly, or shutdown of that system, led to pH
     excursions.

 3.  Instrument Error - the recorder showed an excursion when in fact there
     was none, because the instrument malfunctioned or was out of calibration.

 4.  Instrument Calibration - regular maintenance (i.e., cleaning the
     probe or calibrating the recorder) resulted in a recording of pH
     outside the 6 to 9 limits, when in fact there was no excursion.

 5.  Operator Error - the treatment system failed because of human error.
     (Overcorrection during manual treatment of wastewater is designated
     by Reason Code 2.)

 6.  Diversion in Operation - discharge was either merely blocked or sent
     to a diversion pond or tank as a result of being out of specification
     for treated effluents; but the pH monitor, because of its placement,
     continued to record pH outside the 6 to 9 limits during the diversion
     period.

 7.  Other - the recording of an apparent excursion could be explained by
     a reason code other than those listed in existing codes.  Originally,
     this reason code include those excursions coded as Reason 12, and
     referred to both real and apparent excursions to ensure more detailed
     statistics.

 8.  Unknown - no reasons could be attributed to the occurrence of an
     excursion.

 9.  Emergency Operations - an excursion was uncontrollable for such reasons
     as plant shutdown or power failure.

10.  Spills or Leaks - any spill or leak in any area of the plant, treatment,
     process or general working area which created pH problems that the treatment
     system was not designed to handle.

11.  Rainwater Overflow - because of the fairly frequent occurence of this
     problem, this reason was added to account for those times when excursions
     resulted from treatment system overload from heavy rainfall.

12.  Other - there was an actual excursion that could be explained by a
     reasons other than those listed in existing codes.

 Reason Code II can be divided into "real" and "apparent" groups in the same
 manner as Reason Code I.  For Reason Code II, the apparent reasons (v/hen no
 actual excursion occurs) are Reasons 3, 4, 6, and 7.  The remaining reasons
 (1, 2, 5, 8, 9, 10, 11, and 12) represent observations of excursions that
 correspond to actual  discharges and are therefore considered "real".

                                        23

-------
         4.3  Treatment System Characteristics of Data Base Plants

All plants except one in the combined data base (Data Base I and Data Base II)
use neutralization systems.  The one exception uses ion-exchange technology.  Of
the plants that neutralize their wastewater, five have multistage systems, and
one neutralizes waste only if the wastewater has been diverted.  Two plants
use some form of biotreatment; one has an aeration process; and one uses polymer
additives.  Four plants have settling basins.  Some of the plants have holding
ponds; and two plants have diversion systems in addition to gypsum stacks (ponds),
which can act as redundant diversion systems.  For detailed discussion of BPT pH
treatments systems with respect to industry specific categories the reader is
referred to EPA industry specific Development Documents for Effluent Limitations
Guidelines and Standards.
                                      24

-------
          5.   General  Approach  and  Rationale  for Continuous Monitoring
                               Compliance Criteria


 5.1   Introduction


 As discussed  earlier  in  this document  categorical  standards for  effluent
 limitations guidelines generally required that  the pH  of  plant effluent be
 maintained  in the  range  of  6 to 9,  and some  permits require that pH  be monitored
 continuously.  Although  data indicate  that achievement  of this standard 100
 percent  of  the time is not  realized with Best Practicable Control  Technology
 (BPT)  treatment  systems,  a  standard can  be established  for a percentage of time
 over  which  a  plant must  achieve the required range.

 The lack  of perfect compliance with the  standard of pH  6-9 at all  times is
 in large  part due  to  the  application of  a categorical  standard,  which was
 intended  for  discrete instantaneous sampling (grab or  composite),  to a complete
 enumeration of a time continuum.   That is, a standard  was developed  for use
 with  one  type of sampling mechanism but  applied to more thorough monitoring/
 sampling  requirements.   It  has been and  is Agency  policy  to set  a  standard for
 which  the probability of  violation  at  a  given instant  by  a well  designed and
 maintained  treatment  system is not  greater than approximately .01.   Evidence  of
 this  policy is displayed  in the derivation of effluent  limitations guidelines
 for industrial pollutants based on  the estimated 99th  percentile of  the pollu-
 tant  distribution.  An example of  the  application  of this policy is  exhibited
 in the Inorganic Chemicals  Industry BPT  Development Document.

 When  the  monitoring schedule specifies grab  or  composite  samples,  the
 detection of  a departure  from the  standard is a rare event because the departure
 must  coincide with the time of sampling.   Such  discrete sampling (i.e., grab  or
 composite) provides no information  on  the maximum/minimum value  of the excursion
 or its length  of duration.  With continuous  monitoring, however,  all departures
 will  be detected.  Additionally with continuous monitoring the maximum/minimum
 pH value  and  the length of  duration  for  each excursion  are obtained.  However,
 the introduction of continuous monitoring into  permit requirements necessitates
 a modification of the current pH categorical standard requiring  compliance all
 the time.

 As discussed  earlier, two Agency uniform  data collection  activities  provide
 information on the performance of pH control in several industries.  The six
 plants in the  first Agency  study (DATA BASE  I)  were distributed  among such
 diverse industrial categories/subcategories  as  inorganics,  organics, and iron
 and steel categories, as well as the bauxite, titanium  dioxide and sulfuric
 acid subcategories.   Each of these  plants  provided one year  of strip charts
 from continuous pH recorders.  The  second  Agency data collection activity (DATA
 BASE  II) provided similar continuous monitoring data for  eight inorganic
 chemicals industry plants.  The plants of  the second study  were  distributed
 over the following inorganic chemicals subcategories; aluminum fluoride, chlor-
 al kali, hydrochloric  acid, hydrofluoric acid, hydrogen  cyanide,  sodium bisulfite,
 sodium silicate,  sulfuric acid, and titanium dioxide.   Plants provided 6 to 16
months of pH recorder strip charts.  Data  collected and coded under this activity
were similar in nature to that  of the  first Agency data collection activity.

                                       25

-------
For each of these data bases an excursion is a continuous period when the
plant's effluent is outside the range 6-9.  For each excursion, the date and
time the excursion began, an associated reason code (which identified the reason
for the excursion based on comments obtained from plant logs and/or pH recording
sheets), the maximum or minumum pH values and the total duration time in excursion
were entered into each of the data bases.  The two data bases comprise the infor-
mation used in this report for evaluation of pH standards coupled with continuous
monitoring.

Three fundamental differences exist between the two Agency data collection
activities.  First, the initial data collection activity (DATA BASE I) was
intended to be an examination of continuous pH monitoring across several indus-
tries at the categorical level.  The second data collection activity (DATA
BASE II) covered subcategories within the inorganic chemical industry.  Second,
in DATA BASE II, multiple reason codes were occasionally used to describe
several interacting (i.e., contributing) reasons for an excursion in DATA BASE
II.  Third, as described earlier in this document DATA BASE II utilized a system
with twelve reason codes while DATA BASE I's system consisted of only six codes.

In order to merge the two data bases it was necessary to review those excursions
of DATA BASE II with multiple reason codes and assign a single primary reason
code.  To do this the comment/remark given in DATA BASE II for each multiple
reason code excursion was analyzed arid a decision was made as to which of the
twelve possible reason codes may be considered as the initial reason for the
excursion.

Since in DATA BASE I no code existed for an excursion initiated by operator
error, excursions of this type were reviewed and reassigned a new reason code
based on the explanation provided in the data base.  The results of the
aforementioned receding are shown in Table 1.  The new reason code is called
RCODE on this table.  This table identifies those excursions that have been
reassigned a single DATA BASE II reason code.
                                       26

-------
                                   Table 1

              RCODES ASSIGNED FOR MULITPLE REASON CODE EXCURSIONS AND
                     OPERATOR ERROR EXCURSIONS IN DATA BASE II
RCODES  PLANT  MO.

  10    150    03
  10
  10
150
150
03
03
DAY  YEAR  TIME  PH...  MIN..CODE REASON....   EXPLANATION....

 3   1979  0515    9.3  106  5 OPERATOR ERROR  OPERATOR ERROR
                                               RESULTING IN
                                               RETENTION TANK
                                               OVERFLOW, CAUSED
                                               A MAJOR CLEANUP
                                               AND EXCURSION
                            10 SPILLS OR LEAKS PROBLEM
 3   1979  0710    3.3    3  5 OPEATOR ERROR
 3   1979  0715    9.4    5  5 OPERATOR ERROR  THE REMAINDER OF
                                               THE EXCURSIONS
                                               FROM MARCH 3
                                               THROUGH MARCH 5
                                               RESULT FORM
                                               ATTEMPTS TO
                                               CORRECT TANK
                                               OVERFLOW
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
150
150
150
150
150
150
150
150
150
150
150
150
150
150
150
150
150
150
150
150
150
150
150
150
150
150
03
03
03
03
03
03
03
03
03
03
03
03
03
03
03
03
03
03
03
03
03
03
03
03
03
03
3
3
3
3
3
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
5
5
5
5
1979
1979
1979
1979
1979
1979
1979
1979
1979
1979
1979
1979
1979
1970
1979
1979
1979
1979
1979
1979
1979
1979
1979
1979
1979
1979
0720
0725
0730
0740
0645
0842
0917
0930
0945
1020
1445
2305
0410
0502
0517
0520
0530
0537
0542
0550
0555
0603
0607
0616
0620
0630
2
9
3
9
4
9
9
3
9
10
9
9
9
9
9
9
9
9
9
9
9
9
9
9
9
9
.5
.5
.3
.3
.1
.4
.2
.6
.3
.6
.1
.1
.1
.3
.2
.2
.2
.2
.2
.2
.2
.2
.2
.2
.2
.2
4
3
10
3
4
4
12
5
32
39
3
2
10
11
3
1
3
3
3
3
3
3
3
3
3
3
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
OPERATOR
OPERATOR
OPERATOR
OPERATOR
OPERATOR
OPERATOR
OPERATOR
OPERATOR
OPERATOR
OPERATOR
OPERATOR
OPERATOR
OPERATOR
OPERATOR
OPERATOR
OPERATOR
OPERATOR
OPERATOR
OPERATOR
OPERATOR
OPERATOR
OPERATOR
OPERATOR
OPERATOR
OPERATOR
OPERATOR
ERROR
ERROR
ERROR
ERROR
ERROR
ERROR
ERROR
ERROR
ERROR
ERROR
ERROR
ERROR
ERROR
ERROR
ERROR
ERROR
ERROR
ERROR
ERROR
ERROR
ERROR
ERROR
ERROR
ERROR
ERROR
ERROR
                                         27

-------
                              Table 1 Continued
RCODES  PLANT  MO.  DAY  YEAR  TIME  PH...  MIN..CODE REASON	   EXPLANATION	

                                                   OPERATOR ERROR
                                                   OPERATOR ERROR
                                                   OPERATOR ERROR
                                                   OPERATOR ERROR
                                                   OPERATOR ERROR
                                                   OPERATOR ERROR
                                                   OPERATOR ERROR
                                                   PROCESS UPSET
                                                   OPERATOR ERROR
                                                   PROCESS UPSET   PROCESS UPSET,
                                                                   THAT WAS
                                                                   COMPLICATED BY
                                                                   A PULLED RELAY
                                                                   SWITCH TO TREAT-
                                                                   MENT SYSTEM
10
10
10
10
10
10
10
1

1
150
150
150
150
150
150
150
150

150
03
03
03
03
03
03
03
05

06
5
5
5
5
5
5
5
25

2
1979
1979
1979
1979
1979
1979
1979
1979

1979
0637
0642
0650
0657
0703
0710
0/15
0850

0949
9.2
9.2
9.2
9.3
9.3
9.3
9.3
10.5

10.3
3
3
3
4
3
3
5
9

10
5
5
5
5
5
5
5
1
5
1

1

10

150

782

06

12

2

4

1979

1978

1653

1900

10.6

5.9
5
47 1
5
5 10
   10
782
02
1979  2210
   10
782
03   21   1979  1830
1.9
              5.4
                                                   OPERATOR ERROR
                                                   PROCESS UPSET
                                                   OPERATOR ERROR
                                                   SPILLS OR LEAKS
    11
150 10
         11
     120 10
                                                  2

                                                 11
                                                           SEE 0949

                                                           PROCESS WATER
                                                           OVERFLOW TO
                                                           NORTH SURFACE
                                                           SEWER
SPILLS OR LEAKS PROCESS LEAK TO
                SURFACE SEWER
                COMBINED WITH
                HEAVEY RAINFALL
                SEWER OVERFLOW

SPILLS OR LEAKS NORTH SURFACE
                LINE BEING
                REPAIRED, SIMUL-
                TANEOUSLY LEAK
                IN AN ACID LINE
                INTO NSD CAUSE
                EXCURSIONS, THIS
                WAS COMPOUNDED
                BY HEAVY RAIN
TREATMENT SYSTEM
MALFUNCTION -
                                          28

-------
                              Table 1 Continued
RCODES  PLANT  MO.  DAY  YEAR  TIME  PH...  MIN..CODE REASON....   EXPLANATION	

  10    782    03   22   1979  2015    2.8   40 10 SPILLS OR LEAKS   SEE 3-21
                                                 2 TREATMENT SYSTEM
                                                   MALFUNCTION -
                                                   SHUTDOWN
                                                11
  10    782    03   22   1979  2130    3.0   30 10 SPILLS OR LEAKS   SEE 3-21
                                                 2 TREATMENT SYSTEM
                                                   MALFUNCTION -
                                                   SHUTDOWN
                                                11
  10    782    03   22   1979  2230    2.7   75 10 SPILLS OR LEAKS   SEE 3-21
                                                 2 TREATMENT SYSTEM  MALFUNCTION -

  10    782    03   23   1979  0020    2.8  110 10 SPILLS OR LEAKS,  SEE 3-21
                                                 2 TREATMENT SYSTEM  MALFUNCTION
                                                11

  11    782    07   27   1979  0815    5.1  205  9 EMERGENCY       SPILL  IN
                                                   OPERATIONS      SULFURIC PLANT
                                                                   NORMALLY
                                                                   ADJUSTED IN
                                                                   SURFACE POND,
                                                                   HOWEVER POND
                                                                   WAS OVERFLOWING
                                                                   DUE TP 15"  OF
                                                                   RAIN FROM A
                                                11                 TROPICAL STROM
   6    928    03   11   1979  0415    4.3    1  4                 PH METER WAS
                                                 6                 INCORRECT AND
                                                                   FLOW WAS STOPPED
   6    928    03   11   1979  0440    0.0   25  4                 SEE 0415 SAME
                                                 6                 DATE
   6    928    03   11   1079  0820   13.9   10  4                 SEE 0415 SAME
                                                 6                 DATE
   6    928    03   11   1979  0900    5.6   20  4                 SEE 0415 SAME
                                                 6                 DATE
   6    928    03   11   1979  1010    0.0   20  4                 SEE 0415 SAME
                                                 6                 DATE
   6    928    03   11   1979  1200    5.6  840  4                 SEE 0415 SAME
                                                 6                 DATE
                                         29

-------
After receding the excursions with multiple reasons and/or operator error
excursions a correspondence rule (see Table 3) was developed to convert each of
the 11 remaining DATA BASE II reason codes to an equivalent DATA BASE I reason
code.  Almost all of DATA BASE II's reason codes correspond directly to a
specific DATA BASE I reason code.  The exceptions in the above correspondence
rule occurred for operator errors (DATA BASE II Reason Code 5) and spills or
leaks (DATA BASE II Reason Code 10).  As stated earlier all operator errors
were eliminated from DATA BASE II by reassigning a new excursion reason.
Excursions due to spills or leaks were assigned DATA BASE I reason codes on an
excursion specific basis by utilizing the comments documented in DATA BASE II.
Table 2 displays the spill or leak excursions and the corresponding DATA BASE I
reason code (defined as NCODE) determined from individual review of these
excursions.  Table 3 displays the correspondence rule converting DATA BASE II
reason codes into DATA BASE I reason codes.
                                       30

-------
                       Table 2



SPILLS OR LEAKS ASSIGNED DATA BASE I REASON CODE (NCODE)
NCODES
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
PLANT
102




102




150
150



150
150
150
150
150
150
150
150
150
150
150
150
150
150
150
150
150
150
150
150
150
150
150
150
150
Mo.
11




03




02
03



03
03
03
03
03
03
03
03
03
03
03
03
03
03
03
03
03
03
03
03
03
03
03
03
03
DAY
17




03




10
03



03
03
03
03
03
03
03
04
04
04
04
04
04
04
05
05
05
05
05
05
05
05
05
05
05
YEAR
1978




1979




1979
1979



1979
1979
1979
1979
1979
1979
1979
1979
1979
1979
1979
1979
1979
1979
1979
1979
1979
1979
1979
1979
1979
1979
9179
1979
1979
Time
1720




1025




0108
0515



0710
0715
0720
0725
0730
0740
0745
0842
0917
0930
0945
1020
1445
2305
0410
0502
0517
0520
0530
0537
0542
0550
0555
0603
0607
pH
3.6




3.5




4.7
9.3



3.3
9.4
2.5
9.5
3.3
9.3
4.1
9.4
9.2
3.6
9.3
10.6
9.1
9.1
9.1
9.3
9.2
9.2
9.2
9.2
9.2
9.2
9.2
9.2
9.2
Min.
540




285




17
106



3
5
4
3
10
3
4
4
12
5
32
39
3
2
10
11
3
1
3
3
3
3
3
3
3
Explanation
Excursion from non-contact cool-
ing water discharge outfall.
Resulted from a ruptured hose
in the sulfuric acid unloading
station.
Leak in a cooler recorded in
non-contact cooling outfall.
Failure of pumps to handle
Strom water run-off short cir-
cuiting of organic plant pond.
Drying tower leak.
The remainder of the excursions
from March 3 through March 5
result from attempts to correct
tank overflow.

























                             31

-------
Table 2 Continued
NCODES
2
2
2
2
2
2
2
2
2
3
1
1
1
5
1
1




1
1
1
1
1
1
1
1
1
1

PLANT
150
150
150
150
150
150
150
150
150
150
150
150
150
150
491
491




491
491
491
491
491
491
491
491
491
491

Mo.
03
03
03
03
03
03
03
03
03
03
05
05
05
06
07
07




07
10
01
01
02
02
02
02
02
03

DAY
05
05
05
05
05
05
05
05
05
05
31
31
31
07
05
05




20
11
28
28
06
26
26
26
26
08

YEAR
1979
1979
1979
1979
1979
1979
1979
1979
1979
1979
1970
1979
1979
1979
1978
1978




1978
1978
1979
1979
1979
1979
1979
1979
1979
1979

Time
0616
0620
0630
0637
0642
0650
0657
0703
0710
0715
0630
0718
0755
0124
1850
1950




1645
1100
0425
0445
0055
0945
1050
1130
1220
1225

pH
9.2
9.2
9.2
9.2
9.2
9.2
9.3
9.3
9.3
9.3
9.2
9.2
9.4
3.9
3.0
10.0




0.5
4.8
3.5
4.8
5.4
3.2
3.4
1.2
1.7
2.0

Min.
3
3
3
3
3
3
4
3
3
5
4
7
15
12
60
60




45
10
10
5
10
25
40
30
10
15

Explanation










Leak in steam chest.
Leak in steam chest.
Leak in steam chest.
Acid tank leak.
Cooler leak.
Starting at 1850 2 hours of
excursions resulted from cooler
leak ranging from 3 to 10;
therefore time was divided
equally between the two peaks.
Cooler leak.
Acid leak.
Gasket leak in coolers.
Gasket leak in coolers.
Acid vent overflow.
Acid overflow.
Acid overflow.
Acid overflow.
Acid coolers leak.
Overflow of drying Acid
pump tank
           32

-------
TABLE 2. Continued
NCODES
5
1




1
5
1
1
1


1




1
1
1
1
1
1
1
1
1
1
1
1
1
1
1


1


1

Plant
495
491




491
491
491
491
664


664




664
664
664
664
664
664
664
664
664
664
664
664
664
664
664


664


782

Mo.
03
04




04
04
04
07
01


01




01
01
01
01
01
01
01
01
01
01
01
01
01
01
05


05


12

flay
15
25




25
25
25
21
03


19




19
19
19
19
19
19
19
19
19
19
19
19
19
20
09


27


14

Year
1979
1979




1979
1979
1979
1979
1979


1979




1979
1979
1979
1979
1979
1979
1979
1979
1979
1979
1979
1979
1979
1979
1979


1979


1978

Time
1609
0712




0830
0905
1545
2020
1050


2045




2110
2115
2125
2210
2215
2216
2221
2222
2227
2245
2300
2320
2325
0005
1210


1400


1900

pH
4.5
1.6




2.2
2.4
3.0
3.5
4.4


14.0




0.0
12.0
0.0
0.0
14.0
0.0
14.0
0.0
14.0
0.0
13.0
0.0
14.0
0.0
4.3


10.6


5.9

Min.
10
15




10
30
30
45
1


25




5
10
35
5
1
5
1
5
1
10
20
5
10
90
2


10


15

Explanation
Leak in pump tank.
A leak in the process coolers
through 5:00 PM and attempts
to correct low pH from this leak
accounted for all excursions on
4/25.



Cooler leak.
Acid leak in production being
worked on and acid run over the
mixed trap.
River water screen to drying
tower cooler was being flushed
out of existing acid, and water
went to PVC sewer. Bicarbonate
was added and sewer flushed.














Washing out tail tower and flush
water overflowed curb and entered
storm sewer (HF area).
Caustic spill in blower building
while charging boilers and 1
quart entered drain.
Process water overflow to north
surface sewer.
           33

-------
TABLE 2. Continued
NCODES
1


1



1
1
1
1
Plant
782


782



782
782
782
782
Mo.
02


03



03
03
03
03
Day
04


21



22
22
22
23
Year
1979


1979



1979
1979
1979
1979
Time
2210


1830



2015
2130
2230
0020
pH
1.9


5.4



2.8
3.0
2.7
2.9
Min.
150


120



40
30
75
110
Explanation
Process leak to surface sewer
combined with heavy rainfall
sewer overflow.
North surface line being repaired,
simultaneouly leak in an acid line
into NSD caused excursions; this
was compounded by heavy rain.
See 3-21
See 3-21.
See 3-21
See 3-21.
            34

-------
 Table  3.   Mapping  of  DATA  BASE  II  Reasons  Codes
                           Reason Codes  ( NCODES

     DATA  BASE  II

 1  -  Process  Upset

 2  -  Treatment  System  Malfunction/
     Shutdown

 3  -  Instruement Error

 4  -  Instrument Calibration/Maintenance

 5  -  Operation  Errors  (eliminated
     in screening of multiple
     reason codes)

 6  -  Diversion  in operation but  pH
     monitor still  recording.  (Flow
     stopped or diverted,  but the
     position of the pH probe resulted
     in recording ph of water that  was
     not being  discharged).

 7  -  Other (apparent)

 8  -  Unknown

 9  -  Emergency  Operation

10  -  Spills or  leaks


11  -  Rainwater

12  -  Other (real)
(RCODES (1-12))  into DATA BASE I
(1-6))

     DATA BASE I

1 - Process Upset

2 - Treatment System Upset


6 - Instrument Error

6 - Instrument Error
No mapping - No RCODE
for DATA BASE II
= 5
3 - Technical Excursion -
    Excursion record, but no
    flow to surface waste was
    recorded.
6 - Instrument Error

5 - Unknown

4 - Other

Excursion Specific Mapping
Required

4 - Other

4 - Other
                                        35

-------
5.2  Support of Modified pH Limitations

Having converted and integrated the two Agency data bases, excursions with
reason codes 1 (process upset), 2 (treatment system upset), 4 (other), and 5
(unknown) were grouped and classifed as "real" excursions.  Reason code 3
(technical excursion) and reason 6 ( instrumentation error) were grouped and
defined as "apparent" excursions.  Technical excursions were excluded from
"real" excursions because no discharge of effluent takes place because of
routing of effluent to holding facilities.  Instrumentation excursions were
excluded because a false excursion might be registered due to calibration or
maintenance reasons.

In the discussion and analysis that follows only real excursions are considered.
It is the position of the Agency that apparent excursions (the result of
diversion or instrumentation problems) can be controlled through proper location
of the pH electrode and by maintaining well-operating instrumentation.  A plant
with continuous monitoring requirements has the responsibility of determining
the proper location of the pH electrode so that false excursions due to diversion
are avoided.  Likewise a plant required to monitor pH continuously also has a
responsibility to calibrate, repair, and replace its instrumentation as often
as necessary to provide accurate measurement of pH.

A standard that pH be maintained within the 6-9 range at least 99 percent of
the time is consistent with existing Agency policy setting effluent guidelines.
Thirteen of the fourteen plants in the integrated data base meet the 6-9
categorical  standard at least 99 percent of the time for aggregated months when
analyzing only the "real " excursion times.  That is, a plant's total number of
excursion minutes for a specified number of months in monitoring when divided
by the total number of monitoring minutes for those months is less than or
equal  to .01.  Table 4 displays overall aggregated compliance percerits for each
of the data base plants.

        Table 4.  Number of Months Monitored and Aggregated Percent
                  Monitored Compliance Time for All Real Excursions

                                Number of               Percent Monitored*
Plant                       Months Monitored             Compliance Time

1306                              12                          99.97
2128                              12                          99.99
2653                              12                          99.47
3141                              12                          99.35
6662                              12                          99.71
8011                              12                          99.59
 102                              16                          98.09
 150                               6                          99.40
 491                              14                          99.83
 586                               7                          99.71
 664                               7                          99.91
 782                              12                          99.36
 786                              10                          99.91
 928                               8                          99.996
   - (Number of excursion minutes for number of months monitored)
                total number of minutes monitored.              )

                                       36

-------
Plant 102 displays an aggregate monitored compliance of 98.09 percent.  This
plant had a total of 13427 minutes in excursion over a 16 month period.  Four
individual excursions were extremely long relative to the lengths of other real
excursion times for this plant.  In fact the four excursions account
for approximately 80 percent of the aggregate total excursion time of 13427
minutes.  The four excursions are 2 treatment system upset excursions of 1110
minutes and 2600 minutes, and 2 process system upset excursions of 3360 and
3600 minutes.  These excursions are the reason for this plant's aggregate
percent being less than 99 percent.

Although thirteen of fourteen plants exceeded 99 percent compliance on a
aggregate level, several of the plants did not achieve 99 percent compliance on
a monthly basis  (i.e., non compliance 1 percent or more for a 30 day month is
432 minutes or more).  As displayed in Table 4, percent compliance obtained via
aggregation over a specified duration may provide overall compliance of 99
percent or better.  However examination of percent compliance on a shorter
basis (e-9-. monthly) may reveal compliance substantially less than 99 percent.
For example, consider 2 plants each being monitored for a year and each having
a yearly total of 4800 minutes in excursion.  Assume plant A has 400 minutes of
excursions for each of twelve months.  Based on 30-day months, plant A's monthly
compliances exceed 99 percent  (i.e., 400 minutes in excursion/43200 minutes in
monitoring for a 30-day month equals .9 percent) each month.  Plant A's aggregate
percent also exceeds 99 percent (i.e., 4800 minutes in excursion/518400 minutes
in a year of monitoring based on 30-day months equals .9 percent).  Plant B,
however, experienced all of  its 4800 minutes in one month.  Therefore, its
monthly percent  compliance of 89 percent (i.e., 4800 minutes in excursion/43200
minutes in monitoring for a 30-day month equals 11 percent) is considerably
less than 99 percent although  its aggregate percent still exceeds 99  percent
compliance.  Aggregation of monthly excursions times may provide a distorted
picture of compliance since monthly excursion totals may considerably  exceed
the  1 percent  allowance being  proposed while the aggregate  percent satisfied
the  1 percent  allowance.

Table 5 displays total monthly "real" excursion times for each of the  plants  in
the  integrated Agency data base.  As stated earlier a monthly total excursion  time
of 432 minutes or more exceeds a 99 percent monthly compliance standard.  Eight
of the fourteen  data base plants have at least one month exceeding a  99  percent
monthly  standard.

Plant 2653 exceeds  the  proposed standard in December 1977  (615 minutes)  and
June 1978  (615 minutes).  The  December excursion total  is based on two process
upsets; one  for  195 minutes  and the other for 420  minutes.  The June  excursion
total consists of 4  process  upset excursions of 240, 325, 25 and 25 minutes
duration.  The 325, 25, and  25 minute excursions occurred on the same  day.
Thus, for  plant  2653 the two monthly total  "real"  excursion times are  all the
result  of  process upsets.

For  plant  3141 the  December  1977 total  (1079 minutes),  the  January  (1483  minutes)
and  February  (515 minutes)  1978 monthly  total  "real" excursion times  exceed the
proposed  standard of 432 minutes.   The December total consists of 17  real
excursions of  unknown reason ranging from 2 minutes to  720  minutes  in  duration.
The  January  total of 1483 minutes  is based  on 43 excursions  (2 process upsets


                                       37

-------
 (5,  10 minutes), 6 treatment  system upsets  ranging 5 to 15 minutes, and 35
 unknown reason  excursions ranging from 2 to 330 minutes).  The February total
 of 515 minutes  is based on 30 excursions (3 treatment  system upsets (30, 35, 40
 minutes) and 27 unknown excursion ranging from 3 to 80 minutes).  For each of
 the  months  unknown reason excursions  accounted for almost all of the monthly
 total excursion time.

 For  plant 6662  the November 1978 monthly "real" total  of 690 minutes consists
 of 4 treatment  system upset excursions of 266, 204, 208, and 12 minutes for a 2
 day  period.

 Plant 8011's monthly real excursion total for January  1978 was 2075 minutes.
 This monthly total consisted of 4 real "other" excursions of durations 70, 60,
 45,  and 1900 minutes.

 Plant 102's monthly "real" excursion  totals exceed the proposed 99 percent
 standards in April (3600 minutes), September (4470 minutes), and November 1978
 (540 minutes),  and in March (585 minutes),  June (2823 minutes) and July (849
 minutes) of 1979.  The April total is based on one process upset of 3600 minutes.
 The September total consists of a process upset excursion of 3360 minutes and a
 treatment system upset of 1110 minutes.  The November total  is based on a
 single process  upset excursion.  The March  excursion total is the sum of a 285
 minute process upset and a 300 minute "other" real  excursion.  The June total
 is made up of seven treatment system upsets ranging in durations from 1 to
 2600 minutes.  The July total  of 849 minutes is the sum of twelve treatment
 system upsets ranging in duration for one to 150 minutes in  excursion.  All of
 these excursions occurred in the early part of July.

 Plant 150 exceeded the 99% monthly compliance specification  in March (507 minutes)
 and June (614 minutes) 1979.  The March total  of 507  minutes is based on 66
 "real" excursions, 7 of which were associated with process upsets ranging from
2 to 28 minutes, 46 with treatment system upsets ranging from 2 to 106 minutes,
3 with "other" real  excursions of 2, 2, and 7 minutes and 10 with unknown reason
excursions ranging from 1  to 120 minutes.  The major  reason  contributing to
the March total  was the treatment system upsets.   The June total  of 614 minutes
 is based on  6 process upsets,  13 treatment  system upsets,  and 7 unknown reason
excursions.   The unknown excursions ranged from 3 to  210 minutes, the treatment
 system excursions ranged from 1 to 58 minutes, and the process upsets ranged
from 8 to 57 minutes in duration.

Plant 586 experienced only 2 "real" excursions during the  seven months of moni-
toring data  provided.   The two "real" excursions  were of unknown  reason and
were of 660  and 240 minutes  in duration.

Plant 782 in March 1979 has  a total  monthly "real"  excursion time of 2060 minutes.
The "real" total excursion time is based on 5 process system excursions ranging
from 30 to 120 minutes,  5  treatment system upsets of  20 to 990 minutes, 2  "other"
excursion of 25  and 120 minutes, and 1 unknown excursion  of  15 minutes.
                                       38

-------
Table 5.  Total Monthly Excursion Time By Plant For Real Excursions
                      AGENCY DATA BASE I
Plants:
DEC 1977
JAN 1978
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEPT
OCT
NOV



Plants:
APR 1978
MAY
JUN
JUL
AUG
SEPT
OCT
NOV
DEC
JAN 1979
FEB
MAR
APR
MAY
JUN
JUL
AUG
<. 	
1306
0
0
0
0
120
0
0
0
0
0
2
10



102
3600
0
0
0
85
4470
0
540
0
390
85
535
0
0
2823
849

212!
0
0
0
0
49
0
0
0
0
0
0
0



150









28
127
507
155
127
614


3












AGENCY D/


491


0
165
5
2
30
0
180
80
165
85
245
0
30
45

2653
615
370
120
180
240
125
615
0
120
305
0
120
VTA BASE II


586








0
0
0
900
0
0
0


3141
1079
1483
515
192
13
72
24
3
0
12
20
5



664









238
2
0
0
12
0
14

6662
247
142
0
0
0
293
0
0
0
129
0
690



782




135
110
90
0
29
135
165
2060
145
130
170
220

8011
0
2075
40
45
5
3
0
0
0
0
0
0



786






0
277
75
10
15
0
0
0
0
0















>

928









0
0
0
0
0
0
15
0
                                  39

-------
Of the total 152 months monitored, seventeen (11%) failed to satisfy a 99%
monthly compliance specification and 8 (5%) months would have also failed to
satisfy a 98% monthly compliance.  Slightly less than half of the plants in the
data base (6 out of 14) achieved compliance at least 99% of the time for each
of its months.

The following table classifies each plant's monthly totals of "real" excursion
time with respect to various intervals of monthly excursion percents.  For
example of plant 1306's twelve monthly "real" excursion totals nine months had
excursion percent of zero (i.e. nine months of no "real" excursion time) and
three monthly excursion totals under 1% of a 30 day month (i.e. 3 monthly "real"
totals under 432 minutes).  Clearly within each data base 99 percent compliance
or greater is attainable.  Within Data Base I, sixty-five of the seventy-two
months (90%) have monthly excursion totals under 432 minutes.  Similarly in Data
Base II, approximately 90% of the months have an excursion percent of one oercent
or less.                                                                  K
                                       40

-------
                                 Table 6



Number of Months With(in) The Specified Monthly "Real" Excursion Percent Interval
Data
Base I
Plants
1306
2128
2653
3141
6662
8011
Data
Base II
Plants
102
150
491
586
664
782
786
928
# Months
Monitored
12
12
12
12
12
12
# Months
Monitored
16
6
14
7
7
12
10
8
80
0%
9
11
2
1
7
7
TT
0%
7

3
6
3
1
6
7
33
0%5%
(2161-minutes
or more)






>5%
(2161 minutes
or more)
3






3
                                 41

-------
 Support of Agency policy requiring compliance at least 99% of the time has
 been  documented in an  earlier Agency position paper titled,  pH Under Continuous
 Monitoring.   This work utilized all  excursions in  DATA BASE  I excepT those	
 classified as technical  excursions (Reason  Code 3).  Plant specific data was
 used  for conducting a  simulation study to obtain 99th percentile  estimates of
 total  monthly excursion  times for each plant.  The median  of the  associated
 99th  percent lie monthly  estimates of the  six  plants in DATA  BASE  I  was found to
 be  389 minutes per month in  excursion.  When  previous effluent guidelines have
 been  based on the performance at several  plants,  the median  across  plants has
 often  been used as a basis for a standard.  Following this method,  a standard
 which  allows  one percent of  a month  (i.e.,  432 minutes)  in excursion is
 approximately the same as the median  of 389 minutes of plant specific 99th
 percentile monthly estimates.

 In  PH  Control  of Industrial  Effluents, an industry supported document,  99
 percent  monthly compliance was concluded  to be a reasonable  and attainable
 standard for  plants required  to monitor pH  continuously.   In another industry
 supported study,  Analysis of  Data Collected for EPA By JRB Associates Relating
 to  the  Establishment of  National  pH  Limitations, a monthly compliance standard
 of  98.57  percent  was recommended as  a  result  of the analyses performed.   Thus
 Agency  and industry studies  support  ninety-nine percent monthly compliance as a
 reasonable and attainable standard.

 In  addition to a  monthly compliance  standard  of 99 percent the Agency intends
 to  limit  the  duration  of any  single  excursion,  regardless  of its  pH  value,  to
 30  minutes.   By  reason of the  proposed 99 percent  monthly  compliance  standard
 the Agency acknowledges that  excursions in  the  course  of plant operations  do
 occur.   However  the Agency intends to  limit excessive  individual  excursion
 durations  which  indicate serious  operational  problems  not  rectified  immediately.
 For example,  a  plant might exhibit a single excursion  of 420 minutes  in  a
 given month.   This  plant would  satisfy the  proposed monthly  compliance  standard-
 however,  a single 420 minute  excursion would  certainly be  indicative  of  serious
 plant malfunction not corrected  in a short  period  of time.   Therefore the
Agency has determined that excursions of excessive  duration  should be avoided.
 In conjunction with 99 percent monthly compliance  the Agency position is that
any individual excursion shall not exceed 30 minutes.  The thirty minute limitation
may be shortened  by permit writers when local  water quality conditions require
a more stringent  limitation.   Table 7 displays  the number of real  excursions
on a plant-specific basis that exceed 15, 30 and 60 minutes  duration.   In each
of the Agency data bases approximately 25 percent  of the "real" excursions
exceeded 30 minutes in  duration.
                                     42

-------
                             Table 7

Comparison of Number of "Real" Excursions Exceeding Fifteen (15),
Thirty (30) and Sixty (60) Minutes Duration With Respect to Plant
                          and Data Base
DATA BASE I
Plants
1306
2128
2653
3141
6662
8011
DATA BASE II
Plants
102
150
491
586
664
782
786
928
DATA BASE
I and II
Total
Number of
Excursions
6
2
21
136
13
13
191
28
135
39
2
26
38
7
2
211

Number of
Excursions
> 15 Minutes
3
1
18
33
9
6
70
22
19
19
2
4
24
3
0
93

Number of
Excursions
> 30 Minutes
2
1
15
16
8
4
46
21
11
9
2
2
20
3
0
68

Number of
Excursions
> 60 Minutes
0
0
15
11
8
2
36
18
2
2
2
1
19
2
0
46

              468              163             114              82
                                43

-------
The following table summarizes the earlier tables with respect to 99%
monthly compliance violations and 30 minutes excursion duration violations.
Total violations is the sum of 99% monthly compliance violations and the
30 minute duration violations on a plant specific basis.  For all but plant
928, the number of duration violations exceeded the number of monthly compliance
violations.  This table illustrates the need to restrict individual excursion
lengths in addition to requiring 99 percent monthly compliance.  For example
plant 782 exhibits only 1 monthly compliance violation for the twelve months
of monitoring data provided.  However over half of its real excursions (20 out
of 38) exceeded 30 minutes.  Similar situations occur for plants 2653 and
6662.
                                       44

-------
              Table 8

Plant Month and Duration Violations
        for Real Excursions
# Months
Monitored
DATA BASE I
1306 12
2128 12
2653 12
3141 12
6662 12
8011 12
"7T
DATA BASE II
102 16
150 6
491 14
586 7
664 7
782 12
786 10
928 8
"5D"
152
# Plant Month
Violations For
Monthly Compliance
Specifications
>99%

0
0
2
3
1
1
"T

6
2
0
1
0
1
0
0
TU~
17
# Real
Excursions

6
2
21
136
13
13
191

28
135
39
2
26
38
7
2
WT
468
# Duration
Violations
> 30 Minutes

2
1
15
16
8
4
46

21
11
9
2
2
20
3
0
~68~
114
Total
Violations

2
1
17
19
9
5
53

27
13
9
3
2
21
3
0
78
131
                    45

-------
 5.3   Expected  Value  and  Variance  of  Total  Monthly  Excursion  Time

 This  section presents  another  approach  to  assess plant  specific behavior with
 respect  to  the 99  percent  monthly compliance  limitation.   Total monthly excursion
 times (M) may  be formulated  as  a  random variable obtained  by summing  a  random
 number of individual random  excursion times within  a  month.   By conditioning
 2/JhVUuber  °f excursions  within a month the  expected value E(M)  and  variance
 V(M)  of  the total  monthly  excursion  time may  be obtained.

 Let M =  total  monthly  excursion time obtained by summing N excursion  times
         occurring  within a month.

    N =  total  number of  excursions within a month
           j_ i
    Ti = i     individual excursion time occurring within a month

 Total monthly  excursion  time obtained from N excursions in a  month  may  then
 be expressed as
                            N
    M =  TI + T2 + ...+  TN =  ) Tj


 where N  and the Tj's are random variables.

Assume the individual  excursion times are independent and identically
distributed with mean      and variance  02 .    Also assume the number of
                        T                 T
monthly excursions are independent and  identically distributed with mean
 „  and  variance  Oz  .
  N                 N

The following relationships hold for the expected value and variance of the
random variable total monthly excursion  time (M) when conditioned on the
number of excursions  (N) within a month.

        E(M) =  E (E (M|N))

             =  E (E (T!+...+TN)
                  N    T

        V(M) = V(E(M|N)) + E(V(M|N))

             = V(E(T!+...+TN))  + E(V(T!+...+ TN))

             = V (N v  )  + E (N   Oz )


             =  M2   V(N) +  02   E (N)


             =  y2   a2   +  a2    y
                  T   N       T    N

                                       46

-------
Thus by estimating the mean and variance of the individual excursion times
within months, and the mean and variance of the number of excursions within
months, estimates of the expected value and variance of total monthly excursion
time may be obtained.

Table 9 displays plant specific estimates of the expected value, E(M), and
standard deviation, V(M) 1/2 , of total monthly "real" excursion time.  XM
and S i  represent the arithmetic mean and variance of the number of "real"
excursions experienced each month.  XM and S 2  are estimators of  y
                                             N                      N
 _2 .  XT and S 2 are the arithmetic mean and variance of the duration lengths
  N             T
for the "real" excursions which contribute to total monthly "real" excursion
time.  Each month with no^real" excursion time contributes an excursion time
of zero to the estimators XT and S 2 .  XT and S 2  are used to estimate  M
                                   T             T                         T
and  02 .  By substituting the estimators for  u ,  02 ,  u , and  02  into
      T                                         N    N     T        T
the appropriate expression, estimates of E(M) and V(M) are_obtained.  Estimates
of E(M) and V(M) are identified in the following table as X and S 2 .
                                                                  M
                                       47

-------
       Table 9.  Plant Specific Estimates of Number Of "Real" Excursions
                 Within Each Month, Duration Length In Minutes Of "Real"
                 Excursions Within Each Month, and Total  Monthly "Real"
                 Excursion Time In Minutes
Plant
1306
2128
2653
3141
6662
8011
102
150
491
586
664
782
786
928
Nm*
12
12
12
12
12
12
16
6
14
7
7
12
10
8
XN
.5
.17
1.75
11.33
1.08
1.08
1.75
22.5
2.79
.29
3.71
3.17
.7
.25
S2
N
1
.33
2.39
217.70
2.63
2.45
10.47
517.9
9.87
.57
52.90
11.24
1.57
.5
Nt**
15
13
23
137
20
20
35
135
42
8
29
39
13
9
XT
8.8
3.77
122.17
24.95
75.05
108.4
383.63
11.54
24.57
112.5
9.17
86.90
29
1.67
S2
T
324.31
106.69
12249.61
5369.30
9938.58
178258.88
818642.18
485.89
971.86
55992.86
306.72
26039.15
3562.17
12.5
M
4.4
.64
213.80
282.68
81.05
117.07
671.35
259.65
68.55
32.63
34.02
275.47
20.3
.42
(s2 )
M
15.48
4.78
238.97
443.12
159.83
470.43
1724.39
282.67
93.11
153.14
74.74
409.18
61.76
2.13
*Nm = number of months monitored

**Nt = total number of excursions with length greater than or equal  to zero
  minutes for the specified number of months monitored.   Months with zero
  excursions contribute a single excursion of length zero for each month.

For each plant, the following table displays the number  of estimated standard
deviations of M (S^) which when added to the estimated mean (M) would equal
the proposed limitation of 432 minutes.
     Plant

     1306
     2128
     2653
     3141
     6662
     8011
      102
      150
      491
      586
      664
      782
      786
      928
(432 - M)/SM

    27.6
    90.2
      .9
      .3
     2.2
      .7
     -.1
      .6
     3.9
     2.6
     5.3
      .4
     6.7
   202.6
                                       48

-------
At a minimum, the plants with factors above 3 show that the monthly limitation
of 432 minutes is attainable.  The plants with factors greater than 3 exhibit
no monthly violations of the proposed 99% monthly limitation.  Where previous
effluent guidelines have been based on plant performance, the median across
plants has generally been used as a basis for a standard.  Thus half of the
plants would fall below the limitations and half would be above.  From the
above table approximately half of the plants are in compliance with the proposed
standard.

5.4  Nonparametric Analysis of Number of Monthly Violations

Since plant specific probability distributions of total monthly excursion
times are unknown, the means and variances presented in the previous section
cannot be translated directly into estimates of the probability that a plant
will exceed the standard for a given month.  As an alternative approach, the
following nonparametric formulation is presented.

   Let KI, X2> ••» Xn be a random sample from a Bernoulli distribution, where:

     Xj =1  if a plant's ith monthly total excursion time exceeds the
             monthly excursion standard;

     X-j =0  if a plant's ith monthly total excursion time does not exceed the
             monthly excursion standard.

     n  =    number of months monitored for a specific plant.
             For a Bernoulli distribution, P[X = 1] = e  = 1 - P[X = 0]

Specifically with respect to the proposed total monthly excursion standard the
probability of exceeding the monthly excursion standard is hypothesized to be
.01  (i.e., y = .01).
                 n
Furthermore, S = J"  X-j = a plant's total number of months exceeding the monthly
                i=l
excursion standard.

A  plant's total number of moaths exceeding the monthly excursion standard (S)
has  a binomial distribution such that
              •G)
     PCS = s] =\sja s (1 - b )n"s where s = 0, 1, .... n.

For a fixed n and S = s, a 100 (1- a )  percent confidence interval
for 6 can be obtained.  The confidence interval estimate for e can be constructed
such that the interval satisfies a certain desirable criterion (uniformly most
accurate) and maintains the confidence level  of 100(1- a )  percent (see Testing
Statistical Hypotheses by Lehmann, Sections 3.5, 5.4, 5.5).
                                       49

-------
Various methodologic references have solved the above equations.   (See
Introduction to Statistical Analysis by Dixon and Massey and Experimental
Statistics by M.G. Natrella).                                      	

Therefore, for fixed values of n and s, a 100 (1- a ) percent confidence interval
can be constructed for  B  .  The confidence interval can be used to test the
hypothesis  e =0  , (i.e.,  e = some specified value) with level of signifi-
cance   .  If the specified value  e  is contained in the confidence interval
then the hypothesis is accepted; if  y   is not within the confidence interval
then the hypothesis is rejected.      °

The following table displays plant specific values of n and s and corresponding
90 percent confidence intervals for  Q  = P[X = 1]  = probability the monthly
excursion total exceeds the standard.  Plant specific 90% confidence limits on
 b  for the specified n and s were obtained using Table A-22 from Experimental
Statistics referenced above.

                                            s
                    n             # of Months Standard           90%
Plant      # Months Monitored         is Exceeded         Confidence Interval
1306               12                     0                      0   - .184
2128               12                     0                      0   - .184
2653               12                     2                     .045 - .398
3141               12                     3                     .096 - .500
6662               12                     1                     .009 - .294
8011               12                     1                     .009 - .294

 102               16                     6                     .189 - .619
 150                6                     2                     .093 - .667
 491               14                     0                      0   - .163
 586                7                     1                     .015 - .500
 664                7                     0                      0     .316
 782               12                     1                     .009 - .294
 786               10                     0                      0   - .222
 928                8                     0                      0   - .255

Because confidence intervals can be used to perform hypothesis tests, the
hypothesis  0 = P(X = 1) = .01 can be tested using the preceding confidence
intervals.

The hypothesis  , = P(X = 1) = .01 is rejected for plants 2653, 3141, 102, 150
and 586.  Thus, 9 of the 14 plants fail to reject the hypothesis that the
probability of exceeding the monthly excursion standard is .01.  For plants
2653, 3141, 102, and 586 almost all of the monthly totals exceeding the standard
are due to one or two excursions of considerable duration.  For example, plant
102 had an April 1978 monthly excursion total  of 3600 minutes.  This monthly
total is based on a single process system upset.  Such monthly excursion totals
might be waived if the permittee satisfied requirements specified under the
upset provision of the consolidated permit regulations.  If the 9 plants that
failed to reject the hypothesis of  y = P [X = 1] = .01 are pooled, then for
the aggregate 99 plant months, the monthly excursion total standard is expected
to be exceeded once every 8 or 9 years.

                                       50

-------
     6.  AGENCY RESPONSE AND POSITION ON TECHNICAL CONCLUSIONS SUBMITTED IN
                            SUPPORT OF 1978 PETITION
Technical background and legal precedents are presented in the original petition
("pH Effluent Limitations Guidelines and Standards of Performance for Certain
Manufacturing Point Source Categories, pp. 13-23).  The Agency has reviewed
these conclusions and the submitted supporting evidence in conjunction with
the Agency's data bases and other related information.  The results of EPA's
assessment of major issues raised by the petitioners is summarized below.

Whenever implemented in connection with conventional discrete monitoring
measurement methods, the categorical pH standards, generally the pH range 6 to
9, should be retained as the effluent guideline BPT standards.  In the develop-
ment of effluent standards in general, the Agency has and continues to recognize
that there is no standard that never will be exceeded; therefore, statistical
variations of wastewater characteristics are taken into account in derivation
of guidelines.  Consequently, evidence demonstrating that a standard is attained
for all but a very small percentage of time by appropriate wastewater treatment
systems is not adequate justification alone to adjust effluent standards when
implemented together with compliance measurement methods for which the standards
were developed.  The Agency does recognize, however, that if a standard is
applied along with a requirement for a radically different measurement method
for compliance monitoring then such a requirement may dictate an adjustment of
the original standard.  Consequently, the Agency has concluded that satisfaction
of pH standards whenever final effluent pH is required to be measured continuously
may be beyond the capabilities of BPT and BCT systems.  Standards for pH should
be reformulated.

The EPA independently has verified the petitioners assessment that on an annual,
aggregate basis plants do achieve present categorical pM standards in excess
of 99% of the time.  These findings support the Agency position on the original
standards applied with discrete monitoring.

The Agency concurs with the petitioners that BPT plants with continuous monitoring
can attain categorical pH standards 99% of the time on a monthly basis.  Again,
there  is recognition in the month-to-month variation  in compliance times for
any given plant.  If a plant  is in compliance at least 99% of the time on an
annual basis, it does not follow that the same plant  satisfies the standards
at least 99% of the time for most months.  But for those noncompliance months,
plants experience operational problems repeatedly or  for extended periods of
time.  Such problems include those arising from the industrial process itself,
functioning of the wastewater treatment system, and monitoring instrumentation.
As for malfunctions of the monitoring instrumentation, the Agency maintains
that a permit requirement for continuous monitoring dictates that adequate
instrumentation be installed and that it  be calibrated and serviced routinely.
Furthermore, should a malfunction occur,  it should be  identified and rectified
in a timely fashion.  Agency  data suggest that such maintenance  practices are
not followed at some plants.  For instance, the Agency data base for plant  586
spans a period of seven months of operation.  The pH  strip charts showed 12
occurrences of violations of  the plant pH standard.   As for the  duration of


                                       51

-------
the 12 violations, each of the 4 longest was in excess of 20 hours, and seven
of the remaining 8 were less than half as long.  Everyone of the 4 largest was
due to malfunction of monitoring instrumentation.  The performance of monitoring
instrumentation at this plant relative to other data base plants certainly
raises questions as to whether the plant was meeting the continuous monitoring
requirement.  Although this case may be an extreme illustration, instrumentation
malfunctions are not a factor in modifying pH standards to incorporate continuous
monitoring requirements.

Following the general formulation for effluent guideline standards, there are
a monthly and a daily standard for each regulated parameter.  In this regard,
the petitioners have suggested in various communications a daily standard
formulated as follows:   pH is not to be outside the range 3.5 to 11 for a
continuous period of more than 15 minutes, and pH is not to be outside the
categorical range for more than 15% of the time on a daily basis, which trans-
lates to 3 hours and 36 minutes.  The Agency rejects the concept of a multi-
faceted pH standard for continuous monitoring.  The pH range of 6 to 9 is a
traditional range applied in the practice of wastewater treatment engineering;
whereas, the range 3.5 to 11 appears to be arbitrary.  Furthermore, having two
such ranges would be somewhat inconsistent and would result in a cumbersome
guideline.  The Agency also rejects the 15% daily basis because a daily period
of 3 hours and 36 minutes is excessive relative to a monthly noncompliance
time of 7 hours and 12 minutes and relative to time required to rectify the
potential causes of excursions.  (The petition suggests (p. 22) that most
plants can shut down in approximately 15 minutes.) More importantly, however,
acceptance of this rule would be contrary to the Agency position that the
categorical standards are achievable with discrete monitoring.

Together with the 99% monthly compliance time, the Agency recommends that a
plant is not to be outside the categorical pH range continuously for a period
of more than 30 minutes.  An analysis of this rule is presented in section 5.2
of this document.
                                       52

-------
                                    REFERENCES
 1.    Ayres, G.H.,  Quantitative Chemical  Analysis.  Harper and Row,  New York,
      (1968)•

 2.    Barrow, G.M., Physical  Chemistry.  McGraw-Hill  Book  Compancy,  New York,
      (1966).

 3.    Moore, R.L.,  Neutralization  of  Waste  water  by  pH  Control.  Instrument
      Society of America,  Pittsburgh,  Pennsylvania,  (1978).

 4.    Shinskey,  F.G.,  pH and  pION  Control of  Process  and  Waste Streams.
      John  Wiley and Sons,  New  York,  (1973)."~~

 5.    "An Assessment of pH  Control  of  Process Waters  in Selected Plants",
      Prepared by JRB  Associates,  Inc., McLean, Virginia,  for EPA,  Contract No.
      69-01-3881, Task No.  13,  (March, 1979).


 6.    "pH Control of Industrial Waste  Waters  in the Inorganic Chemicals Industry",
      Prepared by Jacobs Engineering Groups,  Inc., Pasadena, California, for
      EPA Contract  No. 68-01-5167,  Work Order No. 5,  (October, 1979).

 7.    Cleary,  Gottlieb, et. al., Petition to EPA  re:  "pH  Effluent: Limitations
      Guidelines  arid Standards  of Performance for Certain  Manufacturing Point
      Source Categories", Washington, D.C.  (August 3, 1978).

 8.    "pH Control of Industrial Effluent",  Prepared by Centec Consultants
      Inc.,  Reston,  Virginia, for Cleary, Gottlieb, Steen  and Hamilton,
      (June,  1978).

 9.    "Analysis of Data Collected for EPA by JRB Assiciates Relating to the
      Establishment  of National pH Limitations Guidelines", Prepared by
      Centec Corproation,  Reston, Virginia for Cleary, Gottlieb,  Steen and
      Hamilton, (July  1979).

 10.   "pH Under Continuous Monitoring", Agency position  paper.

 11.   "Statistical and  Documentation Support for pH Regulation Development"
      Prepared by JRB Associates, Inc., McLean,  Virginia,  for EPA,  Contract
     No. 68-01-6048, Task  No. 1, (July 1980).

12.  Natrella, M.G., Experimental  Statistics. NBS Handbook 91, U.S. Government
     Printing Office,  Washington,  D.C.  20402 (1966).

13.  Lehmann, E.L., Testing Statistical  Hypotheses.  John  Wiley and  Sons,  Inc.
     New York (1959).
                                       53

-------
                                ACKNOWLEDGEMENTS
This document was prepared by the Environmental  Protection Agency utilizing the
technical and research services of JRB Associates,  Inc.  of McLean, Virginia
(Contract No. 68-01-6048).  Dr. Paul  Cumming was the JRB project director.  The
JRB project manager was Mr. Joseph Ney.  Key JRB staff members contributing to
the project were Mr. Carl  Uhrumacher, Mr. Paul  Campanella, Mr. Rajarshi Ganguli,
and Dr. Charles Norwood.

The latter phases of the study including the preparation of the document were
directed by Dr. Russell Roegner, EPA Project Officer.  The work was supervised
by Dr. Maurice E.B. Owens, Acting Branch Chief for Program Integration and
Evaluation.  Initial technical and critical contributions were made by
Mr. Gary L. Liberson.  Mr. Ray Redd performed important interim data analyses.

The project officer wishes to acknowledge the assistance of the following
personnel at EPA for their contributions in the development of the regulation
and the  supporting materials.  Appreciation is extended to Mr. Richard Gardner
of the Office of General Counsel for his contributions in formulating and
writing  the regulation and the preamble.  Steven Schatzow, now Deputy Assistant
Administrator for Water Regulations and Standards, also contributed to the
formulation of the regulation as a member of the Office of General Counsel.
Mr. Charles Gregg of the Office  of Regulations Review participated actively in
the review and formulation of the regulation and supporting materials, as  did
Mr. Gary Polvi of the  Office  of  Water  Enforcement.   From the  Effluent Guidelines
Division Dr. James Gallup  reviewed and commented on  the regulation at  various
stages of  its development.

Additionally the project  officer wishes  to  express  his appreciation for comments
received from regional  offices.

The project  officer  also  wishes  to thank Ms. Tricia  Mercer of the  Office  of
Analysis and Evaluation for  her  efforts  in  the  typing of  drafts,  necessary
revisions, and for  the final  preparation of the  regulation, preamble,  and
supporting materials.
                                        54

-------
              TABLE CI.A83 5A  (MODIFIED)
  FREQUENCY/PERCENT DISTRIBUTION or EXCURSION TIMES
            FOR REAL REASON CODES BY PLANT
                      PLANT«102

DURATION   FREQUENCY  CUM fREQ   PERCENT   CUM PERCENT
                                 12102 TUESDAY* AUGUST 5» 19BO
       I
       2
       3
       5
      10
      20
      «0
      15
      75
      85
      90
     120
     150
     285
     300
     5«0
    1110
    2600
    3360
    3600
 2
 3
 1
 5
 6
 7
 8
10
12
14
15
18
20
21
?2
23
24
25
26
27
28
 7.143
 3.571
 3.571
 3.571
 3.571
 3.571
 3.571
 7.143
 7.143
 7.143
 3.571
10.714
 7.143
 3.571
 3.571
 3.571
 3.571
 3,571
 3.571
 3.571
 3.571
  7.143
 10.714
 10.286
 17.857
 21.429
 25.000
 28.571
 35.714
 42.857
 50.000
 53.S7I
 64.286
 71.429
 75.000
 78.571
 62.143
 85.714
 09.286
 92.857
 96.429
100.000

-------
              TABLE  CLASS  5A   (MODIFIED)
  FREQUENCY/PERCENT  DISTRIBUTION  OF  EXCURSTON  TIMES
            FOR  RE»L REASON CODES BY PLANT
                      PLANT'150

DURATION   FREQUENCY  CUM  FREQ   PERCENT  CUM  PERCENT
                                            12102  TUESDAY!  AUGUST Sr  1980
       1
       I
       3
       4
       5
       6
       T
       8
       9
      10
      It
      12
      13
      14
      15
      16
      17
      19
      22
      24
      25
      27
      26
      32
      35
      37
      39
      10
      47
      37
      58
      106
      210
32
 9
1
8
22
54
63
74
77
81
84
87
95
99
105
109
110
116
117
118
119
120
121
122
123
124
125
126
127
128
12?
131
132
133
134
133
5.926
10.370
23.704
6.667
6.146
2.222
2.963
2.222
2.222
5.926
2.963
4,444
2.963
0.711
4.444
0.741
0.741
0.741
0.741
0.711
0.74!
0.741
0.741
0.741
0.741
0.741
0.741
0.741
1.461
0.741
0.741
0.741
0.74J
 5.926
16.296
40.000
46.667
54.815
57.037
60.000
62.222
64.444
70.370
73.333
77.778
80.741
61.481
85.926
86.667
87.407
86.148
88.689
69.630
90.370
91.111
91.852
92.593
93.333
94.074
94.615
95.556
97.037
97.778
98.51*
99.259
100.000

-------
              TABLE CL*S3 5A   (MODIFIED)
  FREQUENCY/PERCENT DISTRIBUTION or EXCURSION TIMES
            FOR REAL REASON COOES BY PLANT
                      PLANTM9t

DURATION   FREQUENCY  CUM FREO   PERCENT   CUM PERCENT
                                 12108 TUESDAY* AUGUST S» 1980
       2
       5
      10
      IS
      20
      25
      30
      35
      10
      IS
      60
      90
     ISO
 1
 e
t«
20
21
2S
30
31
32
35
37
38
39
 2.360
17,9/19
20.513
10.256
10.256
 2.561
12.821
 2.560
 2.560
 7,692
 S.126
 2.964
 2,560
  2.560
 20.513
 01,026
 51.282
 61.538
 60.103
 76.923
 79.087
 82.051
 89.700
 90.872
 97.136
100.000

-------
              TABLE  CLASS  5A   (MODIFIED)
  FREQUENCY/PERCENT  DISTRIBUTION  OF  EXCURSION  TIMES
            FOR PEAL REASON COOES BY PLANT
                      PLANT»586

DURATION   FREQUENCY  CUM  FREfl   PERCENT  CUM  PERCENT
     210
     *60
1
2
50.000
90.000
 90.000
100*000
                                12102 TUESDAY,  AUGUST  9r  1980

-------
              TABLE  CLASS  5A   (MODIFIED)
  FREQUENCY/PERCENT  DISTRIBUTION  OF  EXCURSION  TIMES
            FOR  REAL REASON COOES BY PLANT
TUESDAY,  AUGUST 5i  1900
       1
       2
       3
       3
      10
      20
      25
      35
      90
DURATION   FREQUENCY   CUM  FREO    PERCENT   CUM  PERCENT
7
10
11
17
22
23
3
-------
              TAHLE  CLASS 5A   (MODIFIED)
  FREQUENCY/PERCENT  DISTRIBUTION OF EXCURSION TIMES
            FOR "EAL REASON COOES BY PI.ANT
                      PLANT»782

DURATION   FREQUENCY  CUM FREQ   PERCENT   CUM PERCENT
                                           12102 TUESDAY, AUGUST 5, I960
       2
       5
      15
      ZO
      25
      30
      40
      70
      75
      80
      90
      95
     110
     120
     125
     150
     150
     1*5
     205
     270
     990
2
5
7
1
2
1
1
1
4
1
1
2
1
2
1
1
1
1
1
1
I
 2
 7
14
15
17
1"
19
20
24
25
26
28
29
31
32
33
34
35
36
37
38
 5.263
13.158
18.421
 2.632
 5.263
 2.632
 2.632
 2.632
10.526
 2.632
 2.632
 5.263
 2.632
 5.263
 2.632
 2.632
 2.632
 2.632
 2.632
 2.632
 2.632
  5.263
 18.421
 36.842
 39.
-------
              TA"LE  CUA38  5*   (MODIFIED)
  FREQUENCY/PERCENT  DISTRIBUTION or  EXCURSION TIMES
            FOR REAL REASON CODES BY PLANT
                      PlANT«786

DURATION   FREQUENCY  CUM  FREQ   PERCENT  CUM PERCENT
                         12102  TUESDAY,  AUGUST  5.  I960
       2
       5
      10
      15
      60
      75
     210
10.286
14.286
14.286
14.286
14.286
14.286
14.286
 14.286
 28.971
 42.857
 57.143
 7I.42»
 85.714
100.000

-------
              TABLE  CLASS  5A   (MODIFIED)
  FREQUENCY/PERCENT  DISTRIBUTION  OF  EXCURSION  TIMES
            FOR REAL REASON CODES BY PLANT
                      PLANT«926

DURATION   FREQUENCY  CUM  FREO   PERCENT  CUM  PERCENT
       5
      10
50.000
SO.000
 SO.000
100.000
                         12102 TUESDAY* AUGUST 5»  I960

-------
              TABLE Cl*S8 5A  (MODIFIED)
  FREQUENCY/PERCENT DISTRIBUTION OF EXCURSION TIMES
            FOR REAL REASON CODES BY PLANT
                      PLANT»1306

DURATION   FREQUENCY  CUM FREQ   PERCENT   CUM PERCENT
                                           12102 TUESDAY*  AUGUST 5,  I960
       t
      10
      20
      40
      60
I
1
I
1
t
2
3
4
5
6
33.3)3
16.667
16.667
16.667
16.667
 33.333
 50.000
 66.667
 S3.331
100.000

-------
              TABLE  CLASS 5A   (MODIFIED)                    12102  TUESDAY,  AUGUST  5,  i960    10
  FREQUENCY/PERCENT  DISTRIBUTION OF  EXCURSION  TIMES
            FOR REAL REASON COOES BY PLANT
                      PLANT-2128

DURATION   FREQUENCY  CUM FREO    PERCENT  CUM  PERCENT

      IS        1           1       50.000        30.000
      36        J           2       SO.000       100.000

-------
              TABLE  CLASS  5A   (MODIFIED)
  FREQUENCY/PERCENT  DISTRIBUTION or  EXCURSION  TIMES
            FOR  REAL REASON COOES BY PLANT
                      PLANT«S653

DURATION   FREQUENCY .CUM  FREO   PERCENT  CUM  PERCENT
                                 12102 TUESDAY*  AUGUST  3,  I960    11
      10
      IS
      20
      29
      90
     105
     120
     125
     140
     160
     195
     210
     240
     325
     420
 1
 3
 4
 6
 B
 9
12
13
14
15
16
17
19
20
21
I
 4,762
 9.524
 4.762
  .524
  .524
  .762
.286
.762
,762
.762
.762
,762
.524
.762
,762
  4.762
 14.206
 19.044
 28.371
 38.095
 42.857
 57.143
 61.905
 66.667
 71,429
 76,190
 BO.952
 90.476
 95.238
100.000

-------
              TABLE  CLASS 5*   (MODIFIED)                    12102 TUESDAY,  AUGUST  Sr  1980    12
  FREQUENCY/PERCENT  DISTRIBUTION of  EXCURSION TIMES
            FOR REAL REASON CQbES BY PLANT
                      PLANT«Sl4t

DURATION   FREQUENCY  CUM FREO    PERCENT   CUM PERCENT

       27          7        5.107         5.147
       3       2«         Si       17.647        22.794
       5       27         58       19.853        42.647
       6        1         59        0.735        43.382

       7       13         72        9.559        52.941
       8        3         75        2,206        55.147
      10       14         89       10.294        6S.441
      13        2         91        1.471        66.912
      15       12        103        8.824        75.735
      20        3        106        2.206        77.941
      21        2        108        I.471        79.412
      22        3        111        2.206        81,618
      25        t        112        0.735        82.353
      30        8        120        5.882        88.235
      35        2        122        1.471        89.706
      37        1        123        0.735        90.441
      40        1        124        0.735        91,176
      50        1        125        0.735        91.912
      70        2        127        1.471        93.382
      75        2        129        1.471        94.853
      80        I        130        0.735        95.588
      95        1        131        0.735        96.324
     105        1        132        0*735        97.059
     240        2        134        1.471        98.529
     330        1        135        0.735        99.265
     720        1        136        0,735       100.000

-------
              TABLE CUSS 3A   (MODIFIED)
  FREQUENCY/PERCENT DISTRIBUTION OF EXCURSION TIMES
            FOR  REAL  REASON cooes BY PLANT
                      PLANT»6662

DURATION   FREQUENCY  .CUM FREQ   PERCENT  CUM PERCENT
12102 TUESDAY,  AUGUST  5,  19BO    13
      10
      12
      14
      20
      63
      95
     112
     160
     200
     208
     266
     2<»3
1
3
4
5
6
7
e
9
10
11
12
13
T.692
13.3B5
7,672
7.692
7,692
7,692
7,692
7,692
7.692
7.692
7.692
7,692
7.692
23.077
30.769
36.062
16.151
53.846
61.538
69.231
76.923
84.615
92.308
100.500

-------
              TABLE CLASS 5A  (MODIFIED)                    I2|02 TUESDAY*  AUGUST 5,  1980   |4
  FREQUENCY/PERCENT DISTRIBUTION OF EXCURSION TIMES
            FOR REAL REASON CODES BY PLANT
                      PL»NT«80Jt

DURATION   FREOUCNCY  CUM FREO   PERCENT   CUM PERCENT

       1        I           I       7.672         7.692
       2        1           2       7.692        19.385
       52           4      15.385        30.769
      10        1           5       7,692        30.462
      IS        2           7      15.3BS        53.«46
      20        2           9      15.365        69.231
      45        1          10       7.692        76.923
      60        1          li       7.692        64.615
      70        1          12       7.692        92.308
    1900        1          13       7.692       100,000

-------
              TABLE  CLASS  5A   (MODIFIED)
  FREQUENCY/PERCENT  DISTRIBUTION  OF  EXCURSION  TIMES
          FOR  APPARANT  REASON  COOES  BY PLANT
                     PLANTM50

DURATION   FREQUENCY .CUM  FRCQ   PERCENT   CUM  PERCENT
                                                           12102  TUESDAY*  AUOUST  5»
                                                                                           19
       J
       2
       3
       4
       s
       7
      15
      20
      21
      02
      45
     120
 I
 3
 it
 5
 6
 7
 e
 9
IP
11
12
13
15.385
 7.692
 7.692
 7.692
 7.692
 7.692
 7.692
 7,692
 7.692
 7.692
 7.692
 15.385
 23.077
 30.769
 30.462
 46.154
 53,846
 61.536
 69.231
 76.923
 64.615
 92.306
100.000

-------
              TABLE  CLASS 5A   (MODIFIED)
  FREQUENCY/PERCENT  DISTRIBUTION OF EXCURSION TIHES
          FOR APPARANT  REASON  coocfl BY  PLANT
                     PLANTM91

DURATION   FREQUENCY CUM FREO   PERCENT   CUM PERCENT
                                            12102 TUESDAY* AUGUST 5, I960   16
       1
       2
       3
       4
       5
      10
      IS
      ie
      20
      22
      23
      30
      05
      60
      80
     150
     960
    1200
 2
43
 2
 u
26
10
 3
 I
 6
 J
 1
 4
 3
 i
 I
 I
 1
 1
  2
 as
 47
 51
 77
 87
 90
 91
 97
 98
 99
103
106
toe
109
110
111
112
 1.786
38.393
 1.786
 3.571
23.214
 6.929
 2.679
 0,893
 3.357
 0.893
 0.893
 3.571
 2.679
 1.786
 0.893
 0.893
 0.893
 0.893
  1.786
 40.179
 41.964
 45.536
 68.750
 77.679
 60.357
 81.250
 86.607
 87.500
 68.393
 91.964
 94.643
 96.429
 97.321
 96.214
 99.107
100.000

-------
            TABLE  CIA88 SA   (MODIFIED)
FREQUENCY/PERCENT  DISTRIBUTION OF  EXCURSION  TIMES
        FOR APPARANT  REASON  CODES  BY  PLANT
                   PLANT«5B6
12102 TUESDAY? AUGUST  5»  I9BO    IT
DURATION FREQUENCY -CUM FREQ
1 2 2
2 3
5
to
too
1200
1230
1320
1410
4
3
6
7
e
<)
to
PERCENT CUM PERCENT
20.000
10,000
10.000
10.000
10.000
10.000
10.000
10.000
io.ooo
20,000
30.000
40.400
50.000
60.000
70.000
80.000
90,000
100.606

-------
              TABLE  CLASS 3A   (MODIFIED)
  FREQUENCY/PERCENT  DISTRIBUTION Of EXCURSION TIMES
          FOR APPARANT  REASON  COOES BY  PLANT
                     PLANT«66«

DURATION   FREQUENCY .CUM FREO   PERCENT   CUM PERCENT
                                            12102 TUESDAY;  AUGUST 5,  I960   IB
       1
       2
       3
       3
       e
      to
     270
12
 6
 1
 4
 1
 1
 I
12
18
I*
23
21
23
26
46.154
23.077
 3,806
15.38S
 3.S46
 3.646
 3.S46
 46.1S4
 69.231
 73.077
 66.462
 92.306
 96.154
100*000

-------
              TABLE CUA8S 5A   IMODIMEO)
  FREQUENCY/PERCENT DISTRIBUTION  or  EXCURSION  TIMES
          FOR  APPARANT  REASON  CODES  BY  PLANT
                     PLANT»762

DURATION   FREQUENCY  .CUM FREO   PERCENT   CUM  PERCENT
                                                           18102  TUESDAY. AUOU8T 5. 1980   I?
       2
       5
      10
      15
      35
      40
      75
      90
     120
     IflO
     510
     5«0
     780
 2
 4
 6
 7
 8
 9
10
11
12
13
14
IS
16
17
11,765
11.765
11.765
 5.802
 5.682
 5.882
 5.882
 5.862
 5.882
 5.682
 5.682
 5.882
 5.682
 5.882
 11.765
 23.529
 35.294
 41.176
 47.059
 52.941
 56.624
 64.706
 70.586
 76.471
 62.353
 88.235
 94.118
100.000

-------
              TABLE  CLASS  $A   (MODIFIED)
  FREQUENCY/PERCENT  DISTRIBUTION OF  EXCURSION  TIMES
          FOR APPARANT  REASON  CODES  BY  PLANT
                     PLANT«786

DURATION   FREQUENCY -CUM  FREO   PERCENT   CUM  PERCENT
                                 12102 TUESDAY* AUGUST 5r 1980   20
       2
       5
      10
      20
      60
      70
      90
     i«o
     280
     285
     050
     465
     S25
    1785
    1800
 2
 6
 7
 8
 9
10
11
12
13
11
15
16
17
18
19
10.526
21.053
 5.263
 5.26}
 5.263
 5.263
 5.263
 5.263
 5.263
 5.263
 5.263
 5.263
 5.263
 5.263
 5.263
 10.526
 31.579
 36.802
 42.105
 47.368
 52.632
 57.895
 63.158
 68.421
 73.664
 78.947
 84.211
 89.474
 94.737
100.000

-------
              TABLE  CUSS  5*   (MODIFIED)
  FREQUENCY/PERCENT  DISTRIBUTION of  EXCURSION  TIMES
          FOR  APPARANT  REASON  CODES  BY  PLANT
                     PLANT"928

DURATION   FREQUENCY .CUM  FREO   PERCENT   CUM  PERCENT
                                                           12102  TUESDAY!  AUGUST  5»  1980    21
       1
       2
       3
       4
       5
       6
       7
       8
      to
      14
      15
      20
      25
      30
      40
      75
     160
     165
     240
     280
     623
     675
     840
    1800
60
32
 4
 3
22
 3
 2
 3
10
 1
 7
 4
 1
 2
80
nz
116
119
141
144
146
149
159
160
167
171
175
177
178
179
180
181
103
164
165
186
187
188
42.S5J
17.021
2.128
.596
1 .702
.596
.064
.596
5. IP
0.532
3.723
2.126
2.126
1.064
0.532
0.532
0.532
0.532
1.064
0.532
0.532
0.532
0.532
0.532
42.553
59.574
61.702
63.298
75.000
76.596
77.660
79.255
84,574
65.106
88.830
90.957
91.085
94.149
94.661
95.213
95,745
96.277
97.30.0
97.678
98.404
98.936
99.466
100.000

-------
              TABLE  CLASS  SA   (MODIFIED)
  FREQUENCY/PERCENT  DISTRIBUTION OF  EXCURSION TIMES
          FOR APPARANT  REASON  CODES  BY  PLANT
                     PLANT»J306

DURATION   FREQUENCY -CUM  FREO   PERCENT   CUM PERCENT
                         12102 TUESDAY* AU6UST 5, 1980   22
t
2
3
5
7
10
20
30
40
45
SO
60
70
180
229
30
4
3
2
1
1
3
1
3
1
1
2
1
                         229
                         239
                         263
                         266
                         266
                         269
                         270
                         273
                         274
                         277
                         278
                         279
                         281
                         2S2
61.206
10.636
 1.41B
 1.064
 0.709
 0.355
 0.355
 1.064
 0.355
 1.064
 O.ISS
 0,355
 0.709
 0*355
 61.206
 91.804
 93.262
 94.326
 95.635
 95.390
 95.74S
 96.009
 97.163
 90.227
 90.562
 90.936
 99.645
100.000

-------
              TAPLE  CLASS  5A   (HODIFICD)
  FREQUENCY/PERCENT  DISTRIBUTION  OF  EXCURSION  TIMES
          FOR  APPARANT  REASON  CODES  BY  PLANT
                     PLANT-265J

DURATION   FREOUENCY CUM  FREO   PERCENT   CUM  PERCENT
              121
                2
                i
126
127
97.638
 1.573
 0.787
 97.618
 99.213
too.ooo
                                  12102  TUESDAY.  AUGUST  5»  19BO    23

-------
              TABLE  CLASS  5A   (MODIFIED)
  FREQUENCY/PERCENT  DISTRIBUTION  OF  EXCURSION  TIMES
          FOR APPARANT  REASON  COOES  BY  PLANT
                                           12102 TUESDAY* AUGUST 5, 1980   24
       1
       3
       0
       5
       e
      to
      is
      20
      S3
     135
DURATION   FREQUENCY   CUH  FREQ    PERCENT   CUM  PERCENT
3
1
2
I
2
2
t
z
\
i
3
4
6
?
11
13
14
It
It
18
16,667
3,556
11.111
16,667
11.111
11.111
5.556
11.111
5,556
5.556
16.667
22.222
33.333
50.000
fcl.lll
72.222
77.778
08.889
94.144
100.000

-------
            TABLE  CLASS  5A   (MODIFIED)
FREQUENCY/PERCENT  DISTRIBUTION OF EXCURSION TIMES
        FOR APPARANT REASON CODES BY PLANT
                    PLANT*80ll
12102 TUESDAY,  AUGUST  5,  I960    25
DURATION
1
}
5
7
to
15
20
30
35
10
50
65
105
120
160
360
660
FREQUENCY CUM FREQ
49 49
3 52
11 63
2 65
6 Tl










72
73
7«
75
» 77
76
79
60
61
62
2 61
1 85
PERCENT
57.607
1,529
12.941
2.353
7.059
.176
.176
.176
,176
.353
.176
.176
.176
,176
.176
.353
.176
CUM PERCENT
57.647
61.176
71.116
76.171
63.529
64.706
65.662
67.059
66.235
90.566
91,765
92.941
94.118
95.294
96.471
98.924
100.000

-------
i
o
tB
           o
           o    •*>•
           >   irv * e> o- KIK> a o
                                                      rurunirvrvrwxrurururufuivmrururururumfi
                                                                                                                          nifufucu
-------
                         AGGREGATE  PH DATA  8A9E  CONTENTS
                                                                  ilt06  TUESDAY,  AUGUST  5,  1980
                                30URCE«OAT»  PASE  I
DBS
PLANT
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
SI
62
83
84
85
86
87
SB
89
90
91
92
93
94
95
96
97
98
97
100
101
102
103
104
105
106
107
108
1306
1306
1306
1306
1306
1306
1306
1306
1306
1306
1306
1306
1306
1306
1306
1306
1306
1306
1306
1306
13Q6
llofc
1306
1306
1306
1306
1306
1306
1306
1306
1306
1306
1306
1306
1306
1306
1306
1306
1306
1306
1306
1306
1306
1306
1306
1306
1306
1306
1306
1306
1306
1306
1306
1306
MONTH

  2
  2
  2
  2
  2
  2
DAY
       YEAR
TIME
                                                  PH
                  5
                  5
22
22
23
23
27
27
t
2
3
6
8
9
10
15
16
20
20
21
22
23
27
31
9
6
7
10
12
12
13
to
16
17
ia
19
20
22
2«
25
26
27
1
1
2
2
3
3
4
4
4
5
5
8
8
9
H78
1978
i-m
1978
1978
1978
1978
1978
1978
1978
1978
1978
I9?8
19/8
me
1978
1978
1978
1978
1978
1978
1978
1978
1978
1978
1978
1978
1978
1978
1978
1978
1978
1978
1978
1978
1978
1978
1978
1978
1978
1978
1978
1978
1978
1978
1978
1978
1978
1978
1978
1978
1978
1978
1978
1290
1ZOO
1115
H"5
1300
1300
1200
1200
1000
1230
1010
1240
1000
1150
1710 1
910
910
900
940
910
940
940
930
1150
940
1000
920
1000
1000
940
1530
1310
9SO
1010
2400
930
940
1040
945
1020 1
940
940
940
940
1150
1150
920
1140
1140
935
935
1145
1145
1140
».S
5.5
1.
».
•
•
•
t
•
•
•
•
t
•
.
•
t
9
.2
.0
.0
.4
.0
.«
.0
.0
.0
.0
.0
.0
.2
.6
.9
.0
.0
.0
.0
.0
.2
.8
.S
.2
.2
.2
.2
.2
.«
.2
.2
.2
.2
.2
.2
.2
DURATION
RCOOE
NCOOC
                                                           s
                                                          00
                                                          20
                                                          20
                                                          6

-------
                         AGGREGATE  PH DATA BASE CONTENTS
                                                                  II106 TUESDAY,  AUGUST  5,  1980
                                SOURCE»OATA BASE I
OB 3
PLANT
                MONTH
                         DAY
                                YEAR
                                 TIME
                                                  PH
                                                 DURATION
RCOOE
NCODE
109
no
Ml
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
126
129
130
131
132
133
134
135
136
137
136
139
140
141
142
143
|44
145
146
147
148
149
150
151
152
153
154
155
156
157
1S8
159
160
161
162
1306
1306
1306
1306
1306
1306
1306
1306
1306
1306
1306
1306
1306
1306
1306
1306
1306
1306
1306
1306
1306
1306
1306
1306
1306
1306
1306
1306
1306
1306
1306
1306
1306
1306
1306
1306
1306
1306
1306
1306
1306
1306
1306
1306
1306
1306
1306
1306
1306
1306
1306
1306
1306
1306
9
10
10
11
11
12
12
IS
15
15
16
I(>
16
16
IT
IT
19
19
SO
20
22
22
23
23
24
21
25
25
30
30
31
31
1
2
2
4
4
5
12
13
14
15
16
I*
14
20
20
20
20
21
21
24
24
2T
1978
1976
1978
1978
1978
1978
1978
1978
1978
1976
1978
1976
1978
1978
1976
1976
1978
1976
1978
1976
1978
1976
1978
1978
1978
1978
1978
1978
1976
1978
1976
1978
1978
1978
1978
1978
1978
1976
1978
1978
1978
1976
1976
1976
1978
1978
1978
1978
1978
1978
1978
1978
1978
1978
1140
945
945
950
950
945
945
930
1030
1030
920
1030
1040
1040
925
921
900
900
945
945
1050
1050
950
950
1040
1040
940
940
940
940
1000
1000
110
1220
1220
940
940
900
1410
950
1010
940
940
940
940
945
9«5
945
945
1145
1145
1020
1020
1300
4.2
9.2
4.2
9.2
4.2
9.2
4.
5.
4.
10.
5.
5.











1












1















•
,
.
•
•
•
•
•
•
.
•
.
•
•
•
t
•
•
•
•
•
•
•
•
.
t
•
•
•
•
•
•
•
ft
•
•
t
•
•
.2
10.0
2.2
                                                           7
                                                            3
                                                            3
                                                           40

-------
M—• •-v-M iu w iv» (u «J iv» aa uiuo-— a A w

-------
•9


t»
           O

           O
           O
           u
           K


           O
                                                                                        o <\i •> ^ c- e^- cu 4> o m rw » e nj o- o  i\>o in
O   •"


uj   •»
e   <
<   a:
                                                                           oo- o- f*> o v v ov o o o
o o in o o o c>
»w «riv)eoo
o ru — • e e K» e
                                                   o o v> o oir> o r> o e> o o
                                                  .— e> — KI v «r o oeo «r in
                                                   o e o ^ ^ •»«) rurucuo-
o    •
      IU
X    0
a.    cc
      —                                                                ••
tu    O    uj   o- o- c- o- o- c- o- o- e- o- o- e- o- c- e- » e- »• c- 0- o- o- o-
                                                                                        e o opt in in cr> o o m in o o o  o e in in in tn o o o c> ir> intn
                                                                                                   v w — e o — — «« -  ~ — ^ cr e ru nt fu cv rv K> KI in
                                                                                                           o oe> •>•>•> o-  e- •> in i*i«v o- «»~ ^ »• e « &
                                                                                                           •> » «> • oo » •> K>OOO » on a> K> a> « »
                                                                                                     K-^-^»^^-^»^*^^^-^^l>«.r«t»*.^-^»^»fw^|^.f*^.^»
                                                                                                     o- o- c- e- a c- e- c- D- cr- o- ^ o- o- o- e- c- D- » c>- e- cr
c
tu
                                        in « « e- e «.«••««• or~e^.«m « •> tf e- ru
                — — -•— nirufvrv «»i iv w «!«»••»»•»                 _ — —
                                                                                                                                                »~ o- e- o o o
                                                                                                                                                — i.;. ni ru fti


-------
                                           •o-o-a-o-o-o-a-o-o-o-a-a-o-o-d-4'a-a-a.o.o
          uiQi<\ivniMoiu»uiafMMO>4oa.ava-oai<».4anjc»4-*-44aai
          oooooooooooovJiooouioooooooouioavaooooooooovno-ooooovjiui  m
              a a a a a-aa-o-«aaaaaan
-------
•»

i
                                                                            <« iv ni K> tw w o
                                o r>
                           runiirt »u
                                                                                                                                  ooeeeeooirtir t» -, c
                                                                                                                                  o o e> o o o o f« -. — -^. :,'•
                                                                                                                                  O O C> Ki O O P-
cs
C9
           O    IWIXIM
           •>    IT •* »-«.€>• e — ium vt
           e    iy ty iy ruiy mK> KI m mm m m r>> wt v

-------
•0
o
•9
           K

-------
0>
rs
o
CO
hi
          o
          c
                             vt in vi vt vi in vi vi v» vt in in in vi in vi vi v« vi in in vi vi vt in vi in vi vi in v> v> vi vi v> v> v* vi»in v> vi vi vi vi in in v<
                                                               tunivniKinix   inininv>in v « e e

-------
          o
          o   mm AM
          O
          u
t-   AiMmmemmemi
«             «.      .....
a

e
                                                                       em em PI mm o mem tnoo mm e m HI m tnci e o o HI HI o m HI HI
                                                        A> •• • H« HI HI «•• HI ••   »••••• fM»»—   CU«     «•             •»!€«••     ~
CD        it)   v>v> eo e e e me
          x    A»«Air> e AIW w e e
                                                                    e- »• o- »• o- o- » o- c- p-o-o- o- o- c- c- o-
               cr »»-»•»»••«>•»«• e- — —
               AI A» At AI AI AI AI A» AI AI f» f\
                                                                  •• •• AI AI AI AI AI A4 At At AI AI At AI AI AT At AI At AI AI
          O
                                               mininmmin*tn»>in m»> *>»» mm in in mrtminmin in in in «> in in in in m m m in m m m m in

-------
o
o
u
                                                                                                                             n> win»*.*>•*> rf> in in  •>! •>!
o
O

-------
e

o
o
•9
UJ
                                                                     •»• • ««* Ml*
                                                                                                       M ««m At MM III •*•»!«» I* •*(*•*!•»••» <•«••*•«
           UI
           o
           o
           u
           «
           e

           c.
 O
 u
           x
           IT
 o    n
                                                                                          — -— -    — -. M M M fV*. «W M M IV «W«HU M M M — - —
            X

            z
            o
            Z   MMMMMMiU MM MM

            t

-------
O   m»*Ai*»<
o
                       « ni N t»v ««<•
e
o
u
O
o   —MIVIW
z   w« a in in in v> in m »> wt« «««««•»».*>•
O

-------
o
o
m
tu
           O
           u
           o
           O
           u
           ec
           (T
           =9
           O
                              u
                              o
                              O
                              tu
                              O
                              O
                              o
                                    — AI — AI—AIAI—WAIAIAIAlAIAIArAIAtAIAIAIAIAIAIAIAJAI A»~tt Al ll> IT «•>><»• In «>••»*••••• Al Al A«AI AIAI
           «- CV~fUOnCUfVI AIAIAI Al Al Al Al Al Al AIA»«U«VAIA» AI-»ID fUOWM =•—•«)•— — —OOO
                                                                                  — ^   rv a-
                                                                                                                                            r»inis>«K)inot'io
                                                                                                                                            v •» Ate            —
O   —
LJ
     Ul
                                                                    — worufwoe- —
           U.
•-•    A.   oo-ee-me of) v m v i
*^

taj

«c
e    w
I
O.
     III
      Ul
      c    cr
      =>    «
tu    O    u

<
o

E
IS
                 «r
                 »»
                 o-
 O
til
 w
 U
 <£

 O
           o    ni
           o    —
           ar    —
           «    —
           _l    c
           CL    €>
           CD    O
           O    »-
                                      «•    ru *• MM M«      AI At cvi fu AI fw <\i
                                                                                                                                 •»•»•.•»«. lu ru

-------
in


^

•9
          U

          s
          U
          o
          O
r
          o
          x   m«
                                          *•»• o- p- c- e- »• e- €>• o- o- e- e- e- c- c- P- o- e- o- o- «r o-
                                                                                                                        «r « mo- o • c-
                                                                                             o m e «v»- • oin o «r o m o in e e m m m«5«n o
               e 0%iw»- om on •>
                                                                                                                   — •• cw v>tn mmr> ino «>«>
                                                                                                                        •• •• iv «•> ••» ••) «•
          O
          Z   eeo
          «


          a.
                       ooeeoooooooooooeeieooooocoooeeeooccoeoooooeoeoooocee
          e   v
-------
                         AGGREGATE  PH DATA  BASE  CONTENTS
                                                                  11106  TUESDAY, AUGUST 5,  1980   |6
                               SOURCEsOATA  BASE  tl
DBS
PLANT
MONTH
                         DAY
YEAR
TIME
                                                  PH
805
806
807
809
809
810
811
912
913
914
915
816
817
819
819
920
921
922
923
924
925
926
927
829
829
830
831
932
833
834
835
836
837
838
839
940
941
942
843
944
90S
8«6
947
848
949
950
951
95?
953
954
955
856
957
859
150
ISO
ISO
ISO
ISO
ISO
ISO
ISO
150
150
ISO
ISO
ISO
ISO
ISO
ISO
iso
ISO
150
ISO
ISO
ISO
150
ISO
ISO
ISO
ISO
ISO
ISO
ISO
ISO
150
150
150
150
150
ISO
ISO
150
150
150
ISO
150
150
ISO
150
ISO
ISO
ISO
ISO
ISO
ISO
150
150






























5
5
5
5
5
5
5
5
5
5
5
S
5










t
23
23
24
24
27
27
30
30
31
31
31
31
1
4
5
5
7
7
15
1*
16
16
16
17
19
19
19
26
26
29
7
9
9
9
9
10
12
17
17
19
19
25
88
39
31
31
31
31
2
2
4
a
6
7
1979
1979
1979
1979
1979
1979
1979
1979
1979
1979
1979
1979
1979
1979
1979
1979
1979
1979
1979
1979
1979
|979
197?
1979
1979
1979
1979
1979
1979
1977
1979
1979
1979
1979
1979
I9T9
1979
1979
1979
1979
1979
J979
1979
1979
1979
1979
1979
1979
1979
1979
1979
1979
1979
1979
eo5
915
13
850
716
1155
750
900
715
1305
2015
2115
740
1325
750
1450
1650
2050
10)0
2110
2130
2315
?S20
905
905
915
917
715
900
1505
1100
14«5
1550
2130
2157
I3?5
1915 1
900
1152
850
1130
850 !
1057
650 1
215
630
718
755
919 1
1653 1
424
900
12SJ 1C
121 3
.2
.0
.4
.1
.1
.1
.3
.2
.3
.1
.3
.0
.1
.4
.6
.9
.1
.1
.3
.1
.1
.3
.2
.1
.1
.1
.3
.7
.3
.9
.9
.7
.9
.7
.9
.4
.9
.1
.2
.1
.1
,5
.1
.8
.0
.2
.2
,4
.3
.6
.7
,3
.2
,9
DURATION

    3
    2
    1
    «
    2
    3
   29
   27
   13
    2
   13
   10
   10
   42
   24
   14
    9
   3T
    3
    ft
    2
    I
    S
    S
    5
    2
    8
   12
    I
   12
    7
    5
   13
   II
    9
    3
   45
    4
   20
    4
   IS
    V
    I
   13
   16
    4
    7
   15
   10
   47
   57
   IS
    9
   12
RCODE

  3
  3
  3
  I
  0
  9
  I
  I
  I
  I
  I
  t

  3
  I
  I
  2
  2
  2
  2
  2
  2
  2
  8
  2
  2
  2
  I
  4
 12
  2
  7
  2
  2
  2
  2
  3
  3
  3
  I
  2
  I
  t
  1
  1
 10
 10
 10
  I

  I
  1
  1
 10
NCODE

-------
                         AGGREGATE  PH DATA  BA3E  CONTENTS
                                                                  1H06 TUESDAY,  AUGUST 5«  1980    it
on s
PLANT
                MONTH
DAY
                        90UHCE"DATA BASE II —

                         YEAR    TIME      PH
DURATION
859
860
861
962
963
864
865
866
867
969
869
970
971
972
973
974
975
976
977
979
979
880
991
982
883
984
885
886
887
888
8*9
890
89|
892
893
894
895
996
897
198
89?
900
901
902
903
904
90S
906
907
909
909
910
911
912
ISO
150
150
150
150
ISO
150
ISO
ISO
ISO
ISO
ISO
ISO
ISO
ISO
ISO
ISO
ISO
150
ISO
ISO
491
191
491
491
«91
49|
191
4?1
491
491
491
491
491
491
191
491
491
191
491
491
491
• 91
491
491
491
491
491
491
491
091
491
491
491
























T
7
T
T
7
7
7
7
7
7
7



















13
IS
15
IS
IS
16
19
10
IS
IS
18
19
19
19
19
19
20
20
21
21
26
29
29
29
5
S
10
IS
IS
16
16
16
16
20
20
22
4
4
S
25
4
l«
16
21
21
21
22
21
25
25
25
25
25
28
1979
1979
19f9
1979
1979
1979
1979
1979
1979
1979
19T9
1979
1979
1979
1979
1979
1979
1979
1979
1979
1979
1979
1979
1979
1979
1979
19T9
1979
1979
1979
1979
1979
1979
1979
1979
1978
1979
1978
1979
1979
1979
1978
1979
1979
1979
1979
1978
1978
1978
1978
I9re
1978
1978
1978
M?0
710
810
838
1027
1155
935
lots
tO«7
1320
1435
1915
2015
727
1100
1830
915
1319
1112
1515
I2«0
600
1735
700
1850
1950
2200
2030
?32S
215
250
615
708
930
J6«15
1930
2*?5
23«5
720
230
1550
720
700
1330
1630
1830
?iSO
1945
1950
2000
2015
2020
2120
7ld
9.1
9.2
5.6




















.0
.1
.2
.5
.7
.7
.3
.2
.6
.5
.9
.1
.5
.1
.1
.1
.1
.1
.6
.9
5.6
3.0
10.0
5.2
10.0
10.0
10.0
10.0
1.9
10.0
1.8
o.s
10.0
5.7
5.7
10.0
10.0
5.4
S.9
5.9
5.9
5.5
5.4
1.5
10.0
1.1
1.9
2.5
5.5
5.3
5."
8
12
2
58
35
3
10
22
2
10
1
12
210
3
15
40
11
1
1
4
3
IS
4S
OS
60
60
60
30
15
10
5
23
22
to
45
10
5
5
S
30
15
2
2
2
2
2
10
2
2
2
2
2
10
2
RCODE

  2
  I
  2
  2
  2
 II
  2
  2
  2
NCODE

  2
  I
  2
  2
  2
                                                                     1
                                                                     1
                                                                     1

-------
a a a a a e a
                                                -a-« •« •» -a -o « o -o -a -a -o -a -o -a -o   •>
                                                                                           a

                                                                                           •^
                                                                                           x
                                                                                                      n
                                                                                                      CT
                                                                                                      o
                                                                                                <•    »
                                                                                                O    -I
                                                                                                e    m
                                                                                                a
                                                                                                r»    TI
                                                                                                rn    x
                                                                                                • i
                                                                                                o    a
                                                                                                OJ

                                                                                                »
                                                                                                ca
                                                                                                •»    c
                                                                                           O

                                                                                           a
                                     v« vi ni «» n> 

                                                                                           9

-------
AGGREGATE PH DATA BA«E CONTENTS
     11106 TUESDAY,  AUOU8T 5*  I960    14
OB 9
967
966
969
970
971
972
973
974
975
976
977
978
977
780
781
9fl2
983
964
985
966
987
788
787
770
771
772
993
774
775
776
97 T
998
779
1000
toot
1002
1003
1004
1005
1006
1007
1006
|009
1010
ton
1012
1013
1014
1013
1016
1017
1016
1019
1P20
PLANT MOM
491 12
49| 12
491 12
491 12
491 12
491 12
49| (2
491 12
491 12
491 12
491 12
491 12
491 it
491 11
491 I!
491 I!
491 12
091
491
491
491
491
471
471
471
491
491
491
491
491
471 !
471 J
471 <
471 2
471 i
491 i
491 !
491 I
471 !
491 3
491 I
491
491
491
491
491
491
491
49t
491
491
491
491
491
TH DAY
11
II
It
II
II
II
II
II
II
II
> 11
> II
' II
' 11
' 11
• 11
' It
13
14
16
19
21
21
21
21
26
28
29
29
29
» 6
'. 26
! 26
'. 26
> 26
! 26
» 26
I 4
1 4
I 3
\ 7
7
8
12
14
13
16
16
19
22
25
6
7
25
      80URCE»DATA BASE  II
YEAR

1978
1978
1978
1976
1978
1778
1978
1976
1976
1978
1978
1978
1978
1978
1978
1978
1978
1979
1979
1979
1779
1979
1979
1977
1777
1777
1979
1979
1979
1979
1979
1979
1979
1979
1979
1979
1979
1979
1979
1979
1979
1979
1977
1779
1979
1979
t979
1979
1979
1979
1979
1979
1979
1977
TIME

 850
 935
 903
 950
1110
1135
1112
1200
1215
1230
1250
JSflO
I4?0
2031
2033
2047
1500
teoo
1340
1740
 700
1420
1450
rooo
 425
 045
1645
1703
1710
  55
 943
1050
1130
1220
1230
1345
 930
 935
 715
 745
 747
1225
1015
2200
1603
 700
1740
tooo
 620
1715
1700
                71?
 PH

10,0
 5,3
10.0
 3.6
 9.1
 4.6
 4.7
 4.8
 5.6
 3.9
 9.7
 «,6
 4.6
 3.6
 0.3
 2.8
 4.0
 2.4
 5.4
 4.6
 0.0
 3.6
 • .8
 3.2
 9.6
 3.3
 4.8
 5.6
 5.7
 9.7
 3.4
 3.2
 3.4
 1.2
 1.7
10.0
10,0
 2.4
 9.«
 5.2
 5,4
10.0
 2.0
 3.4
 3.7
 4.3
10.0
 5.2
10.0
10.0
 5.6
 5.8

 U6
                               DURATION

                                   5
                                   5
                                   5
                                   3
                                   t
                                   3
                                   5
                                  10
                                   9
                                   3
                                  30
                                   5
                                  10
                                   1
                                   3
                                   5
                                   5
                                  10
                                   5
                                   3
                                  30
                                   2
                                  10
                                  20
                                  20
                                  10
                                   3
                                   3
                                   3
                                   3
                                  10
                                  25
                                  40
                                  30
                                  10
                                  20
                                  30
                                   3
                                  10
                                   3
                                   2
                                   2
                                  15

                                  45
                                  10
                                   3
                                   5
                                 130
                                  10
                                   2
                                   5
                                   5
                                  13
RCODC

  T
  7
  T
  T
  7
  7
  7
  7
  7
  7
  7
  7
  7
  7
  7
  7
  7
  4
  4
  4
  7
  4
  2
  2
  2
 10
 10
  I
  2
  2
 10
 10
 10
 10
 10
  2
  I
  7
  r
  7
  4
  4
 10
  4
 12
 10
  2
                                            NCODE
  10

-------
                          AGGREGATE  PH DATA BASE CONTENTS
                                                                 ItlOfc TUESDAY, AUGUST 5, I960   20
                                SOURCE-DATA BASE  II
DBS

1021
1022
102)
1024
1029
1026
1027
I02B
1 029
1030
1031
1032
1033
1031
I03S
!03«
1037
1058
1039
1040
I04|
1042
104)
1044
1049
1046
1047
I04S
1049
1050
1091
1092
1093
1094
1099
1096
1097
1098
1099
1060
1061
1063
1064
1069
1066
1067
1068
1069
1070
1071
1072
1073
1074
PLANT

 491
 491
 491
 491
 491
 491
 491
 491
 491
 491
 986
 986
 986
 986
 986
 986
 986
 986
 986
 986
 986
 986
 664
 664
 664
 6*4
 664
 664
 664
 664
 664
 664
 664
 664
 664
 664
 664
 664
 664
 660
 664
 66"
 664
 660
 664
 664
 664
 664
 664
 664
 664
 664
 664
 660
MONTH
DAY
YEAR
PH
4
4
4
4
4
4
4
4
6
7
2
2
2
3
3
4
4
9
6
6
6
6
































29
29
25
29
29
29
25
25
14
21
7
12
14
19
18
8
9
24
19
22
22
22
3
3
4
4
4
4
8
10
It
II
It
12
15
19
19
19
19
19
19
!«
19
19
19
19
19
19
19
19
19
19
19
|9
1979
1979
1979
1979
1979
1979
1979
1979
1979
1979
1978
1978
1979
1979
1979
1979
1979
1979
1979
1979
1979
1979
1979
1979
1979
1979
1979
1979
1979
1979
1979
1979
1979
1979
1979
1979
1979
1979
1979
1979
1979
»979
1979
1979
1979
1979
1979
1979
1979
1979
1979
1979
1979
t979
730
830
840
050
905
935
1309
1515
1000
2020
1215
1230
2150
2200
2100
1800
1800
1400
1345
6Q5
700
900
)045
1050
2130
2145
2159
2230
1135
?250
1050
1120
1345
910
1245
1830
1840
1900
1905
1935
1945
1950
2000
2045
2110
2113
2125
2210
2215
2216
2221
2222
2227
2245
10.0
2.2
».6
10.0
2.4
10.0
9.8
3.0
9.2
3.9
4.8
3.7
9.6
5.7
9.9
9.9
5.6
9.1
5.2
10. Q
0.0
9.6
».4
4.4
1.2
.9
.7
.8
.5
*7
*t
.1
.6
.4
.9
.3
.0
.6
.9
2.9
9.2
13.4
14.0
14.0
0.0
12.0
0.0
0.0
14.0
0.0
14.0
0.0
14.0
0.0
DURATION

    39
    10
     5
    19
    30
    90
    19
    30
    30

    'I
     2
     1
   660
   240
  1230
  1320
   100
  1200
    10
     9
  1440
     !
     I
     1
     1
     •
     t
     9
     1

     I
     2
     9
     I
     2
     2
     9
    10
     3
     1
     1
     I
    25
     5
    10
    35
     9
     1
     5
     t
     9
     1
    10
                   RCODE

                     2
                    10
                     I
                     2
                    10
                     2
                     t
                    10

                    .5
NCODE
                                                    I
                                                     3
                                                     3
                                                    10
                                                    10
                                                    10
                                                    10
                                                    10
                                                    10
                                                    10
                                                    10
                                                    10
                                                    10
                                                    to

-------
WMMWMMWIMIWM
                                                                                                      nirurun*   rw^
                               «»nj M niM» •*MM«*ra
                                                                                                                 •a -a J» o   a   n   -a
                                                                                 •>4<«M.Bnta»»-«on»niMn»o>
                                                      ~» out— u»—
                                                                                                      va o u ui o m mo o
                                                                                                                        M   -o   M   O
^AM 9 «•««« •««•«• W-4-0 MO
   oooi OOOIM ooo una«ioiivi««i««iv(li«»oioomoiwiouioui«ji«io —K«OO«»O M«»UI«*«» iwm — —(woo
       MM MIN •.»>—«»«•»•>* 01 VH\MUI» »»» MM M

-------
O
oooeom
           IT

                         ~ -• »^ KI nj — ». o «n ru « KI e- KI ir>

           z   ru fu ru ru •
                rwKitnm m KI M



-------
                                                                                                       tu iu rw ni M ru n> lu n> M* » •»
                                                                           o ^ ^ -a w n« (M i\» — —
-a -a -o ^j -o -o j -o -o -o -o -o  -a -a -o -o -o « -o  -a  o -o -a o -J -o -o -o  -o -o -a -o ^ -o -o -u -a -a -o -o -a -a -o o -o -a < -o -o
                                                 vouiiflooooouiouiooot(»ruoooovflr«oo«— oooom — o o o — o    m    9    ai
                                                                                                                                             — a-o a    -a    «-•    a
                                                                                                                         OIUOOOOUIQOOO

-------
crutui ouimo oouimo moo VIMO v«o MOMVHOOO u«o   m   ta    or
                                                                      o
                                                                      a
                                                                               C0
                                                                               o

-------
til
     o
     o
     u
                                                                                                                                                          I 0-0
           Ul
           o
           O
           u
           o
                  . — ...ofwoiv
                              ru
                                                                                         •   tu   >•    (u
                                                                                                                                                      o e
                                                                                                                                                      o —
09

*•
o
                               •  o o -* o oo-»< e i
                                                                                                                                              — IflfWC—
UJ
03
UJ
CO
     x    0 in cvni
cjootr>
m — cw cr i
if>iroiM«ri^o
o er 9 9 «r ruin
                                                                                                                              ooru*inir.
                                                                                                                              «• inm m in o
                                                                                                                                                              tn
                                                                                                                                                              
ts
           o   -. —• — — iv iv ru ru ru M IM ru iv ru nnv ru iv iv f« ^
           O


-------
                  fl u ui um ui mui ui «i a a A a
iM M M nt nj iw m m ni nj iu n« ni iu nj (u iu nj ru iu nt    a
      — »-< -4 » a .o a a  .» iviru ru  -a -a -a .a    m
*W'MM<>^<«4'^^^-^«4^-^-^^-^-i4-^>^^^4«4    »
              4 -O « -O -O O -4 -O-O 4 4 -» y> 4 -O -O
•o at aa UIWAI
                    o VMO ni o o 4 •• o » uiui-4ui
        sa ui o o ui M ui m nj nt •• ui •- va«—
                                                                          e*
                                                                          o
                                                                          a

-------
                          UNITED  STATES
                 ENVIRONMENTAL PROTECTION AGENCY
                    BEFORE THE ADMINISTRATOR
In Re:
     pH Effluent Limitation Guidelines and
     Standards of Performance for Certain
     Manufacturing Point Source Categories
                PETITION FOR REVIEW AND REVISION
          Submitted herewith is a report entitled "pH Control of

Industrial Effluents" prepared by Centec Consultants, Inc.  ("Cen-

tec Report").  The Centec Report demonstrates conclusively the

impracticality of the current regulations and policies of the

Environmental Protection Agency concerning the control of pH

in effluent discharge for certain point source categories.

Petitioners hereby request the Agency to promptly review these

regulations and policies and amend them so as to establish prac-

tical and realistic parameters for pH control.

          This petition and evidence presented in the Centec

Report establish the need for the revision of certain of EPA's

pH effluent limitations guidelines for existing plants and standards

of performance for new sources promulgated under the authority

of Sections 301(b), 304(b) and 306(b), respectively, of  the  Federal

-------
Water  Pollution  Control  Act  ("the Act"), 33 U.S.C. §§ 1311(b),



1314(b)  and  1316(b).   This petition  is submitted under authority



of  Sections  301(b),  304(b) and 306(b) of the Act and 5 U.S.C.



§ 553(e) which grants  to "an  interested person the right to



petition for  the  issuance, amendment or repeal of-a rule."  Peti-



tioners own  and/or operate industrial plants which are point sources



subject  to these  regulations  and are therefore "interested persons"



within the meaning of  this section.




          This petition  is submitted pursuant to the procedures



detailed by  the United States Court of Appeals for the District



of  Columbia  in Oljato Chapter of the Navajo Tribe v. Train, 515



F.2d 654, 666 (D.C. Cir.  1975) as the most direct and expedient



method of bringing Petitioners' findings and conclusions before



the Agency and requesting the Agency to initiate the necessary



regulatory revisions mandated thereby.  See also Simon v. Eastern



Kentucky Welfare Rights Organization, 426 U.S. 26 (1975)(dissent-



ing opinion of Brennan, J.);  Save the Bay,  Inc. v. Administrator



of EPA, 556 F.2d 1282  (5th Cir. 1977); Ethyl Corp. v.  Environ-



mental Protection Agency, 541 F.2d 1 (D.C.  Cir. 1976).



          Specifically, Petitioners hereby petition the Agency



to promptly take the following actions:



          (1)  Rescind the Agency's internal policy limiting the



pH of effluent discharge to the range 6.0 to 9.0 on a continuous



basis for industrial point source categories involving the manu-



facture,  or use  in the manufacturing process,  of strong acids or



substantial quantities of bases,  and
                               — 2 —

-------
          (2)  Revise the pH parameters in effluent limitations

guidelines for existing plants and standards of performance for

new sources, or issue new parameters where no regulations are

currently effective, for point source categories involving the

manufacture, or use in the manufacturing process, .of strong acids

or substantial quantities of bases, to permit excursions outside

the pH 6 to 9 range for a total period of at least one percent

of a month,~  with excursions below pH 3.5 or above pH 11 being

limited to 15 minutes per excursion.  These categories and sub-

categories  (hereinafter referred to as "the specified subcategories")

include but are not limited  to the following:

     A.   Existing Guidelines

          1.   Inorganic Chemicals (40 C.F.R. Part 415):

               Subpart D  (Calcium Chloride)

               Subpart F  (Chlorine and Sodium or Potassium Hydroxide)

          2.   Fertilizer Manufacturing  (40 C.F.R. Part  418):

               Subpart B  (Ammonia)

               Subpart C  (Urea)

               Subpart D  (Ammonium Nitrate)

          3.   Iron  and  Steel Manufacturing  (40  C.F.R.  Part  420)
 I/    Certain  types  of  plants may  require  a  greater  than  1%
 ~    excursion allowance.   The  Phosphate  fertilizer manufacturing
      category, for  example,  is  not  addressed  by this Petition.
      That industry  has indicated  in the past  that it could  not
      meet a 99% monthly compliance  requirement.  Some plants in
      the categories covered by  this petition  may not be  able to
      meet that level of performance because of problems  unique
      to that  plant. See p.  22  below.
                                -3-

-------
     4.   Non-Ferrous Metals Manufacturing (40 C.F.R. Part 421)

          Subpart A (Bauxite Refining);

          Subpart B (Primary Aluminum Smelting)

          Subpart C (Secondary Aluminum Smelting)

     5.   Phosphate Manufacturing (40 C.F.R.  -Part 422)

          Subpart A (Phosphorous Production)

          Subpart B (Phosphorous Consuming)


B.   Categories Without Regulations But
     Subject to Agency "Policy" on pH

     1.   Inorganic Chemicals Manufacturing,  various
          subparts including:

          Subpart G (Hydrochloric Acid)

          Subpart H (Hydrofluoric Acid)

          Subpart J (Nitric Acid)

          Subpart 0 (Sodium Carbonate)

          Subpart U (Sulfuric Acid)

          Subpart V (Titanium Dioxide)

          Subpart W (Aluminum Fluoride  Production)

          Subpart Y (Ammonium Hydroxide)

          Subpart AP (Hydrogen Cyanide)

          Subpart AV (Strong  Nitric  Acid)

          Subpart BF (Sodium  Silicofluoride Production)

     2.    Organic Chemical  Manufacturing  (40  C.F.R.  Part  414)

          Subparts covering plants manufacturing, or using

          in the  manufacturing process, strong  acids or bases.

     3.    Fertilizer Manufacturing (40  C.F.R.  Part  418)

          Subpart E (Nitric Acid)
                          -4-

-------
          The necessity for these Agency actions  is  documented



in the Centec Report which extensively analyzes the  operating



histories of the efforts of many different types  of  plants  to



achieve pH control.  The purpose of this petition is to summarize



and place in perspective the technical data and other information



contained in the Centec Report.








I.   PETITIONERS AND THEIR INTEREST



          Petitioners herein are:



               Allied Chemical Corporation



               American Cyanamid Co.



               BASF Corporation



               C F Industries, Inc.



               E. I. du Pont de Nemours & Co.



               FMC Corporation



               Hooker Chemical Corporation



               Kaiser Aluminum & Chemical Corporation



               Olin Corporation



               Union Carbide Corporation



               United States Steel Corporation



These companies produce a diverse range of products including



organic and  inorganic chemicals, fertilizers, iron and steel,



non-ferrous metals and phosphates.  Each is vitally affected by



the Agency's regulations and policy limiting pH in effluent dis-



charge, as  it owns or operates one or more plants which discharge



effluent pursuant to a National  Pollution Discharge Elimination
                                -5-

-------
System (NPDES) permit containing pH parameters.   Petitioners'



common interest in filing this petition is in presenting to



the Agency the new data and other new information contained



in the Centec Report demonstrating the impracticability of certain



industry subcategories complying with current EPA -regulations



and policy requiring that pH be maintained within the narrow



6.0 - 9.0 range 100 percent of the time.








II.  BACKGROUND



     A.   pH As A Pollutant Parameter



          pH is a value which measures the acid  (concentration



of H+ ions) and alkaline (concentration of OH- ions)  content of



solutions.  From a neutral pH of 7 the addition  of hydrogen ions



will lower pH while the addition of hydroxyl ions will raise it.



Since pH measurement is logarithmic, at high or  low pH values,



i.e., high concentrations of H+ or OH- ions, the addition of



relatively large amounts of acid or base  is required  to change



pH, while at pH values close to neutral the addition  of extremely



small amounts of acid or base will produce large chancges in pH.



It is this latter characteristic of pH that makes its contin-



uous regulation within a narrow range around the neutral point



of 7 very difficult.  (Centec Report §§ 2.1, 2.3.)






     B.   The Initial pH Regulations



          The 1972 amendments to the Act,  which  extensively



revised the mechanism under which all effluent discharges from
                                -6-

-------
industrial plants and municipalities are regulated,  required
achievement by all point sources of Best Practicable Control
Technology (BPT) by 1977 and Best Available Technology (BAT)
by 1983.  When effluent guidelines for BPT were first proposed
in 1974, the almost universal standard was to maintain control
of pH between 6.0 to 9.0 standard units.
          In the absence of actual experience in maintaining
pH within such a narrow range, this requirement was generally
accepted by industry, and equipment designed to control pH at
this level was installed by many major corporations, including
several of the Petitioners.  Previous experience with pH control
had been with grab and/or composite samples; essentially no data
on continuous monitoring was then available.  As these systems
began operating with continuous pH monitoring, however, it became
clear that many factors not readily apparent from theoretical
considerations caused the control of pH between 6 to 9 at all
times to be extremely difficult,  if not impossible, for some
plants.
          The effluent guidelines for the  fertilizer industry
produced the  first legal challenge  to EPA's pH regulations.
The nitric acid regulations promulgated April  8, 1974, included
a requirement for control of pH between 6  and  9 on  a continuous
basis.  The industry asked  for  some modest relief from this require-
ment.   Because  of the lack  of data  on pH  control, both the Agency
and industry  retained consultants to  study the technical  aspects
                                 -7-

-------
of pH control.  The  industry hired Centec Consultants, Inc. while

EPA commissioned Jacobs Engineering to perform a study.

          The Centec Report submitted to EPA on June 18, 1976, con-

cluded that it was neither feasible, nor was it desirable from the

point of view of resource conservation, to attempt to control pH

between 6 and 9 at all times.  Centec recommended that nitrogen

fertilizer industry pH levels be controlled between 6 to 9 for

99 percent of the time on a long-range average basis and that
                                                       I/
brief excursions outside of a wider range be permitted.

          The Jacobs Engineering Report also concluded that it is

not possible to control pH between 6 and 9 for 100 percent of
         I/
the time.    On March 25, 1977, EPA revoked the pH regulations
                                                       $/
for nitric acid and phosphate fertilizer manufacturing.    The

Notice of Revocation stated in pertinent part:
I/   The basis for this recommendation was (1) the extremely high
     cost required to remove most of the last few "blips" of pH
     outside 6 to 9 in the effluent, (2) an analysis that showed
     that no significant hydrogen ions would be discharged to the
     waterway until a pH of 3.5 was reached, (3) the results of
     studies at du Font's Haskall laboratories that showed that
     at a pH of 3.5, fish showed little effect from pH exposure
     for short periods, and (4) the impossibility of removing all
     chances of excursions from the pH range of 6 to 9.
          Relevant portions of this June 18, 1976 Centec Report
     and the pH control section of a September 15, 1976  Centec
     Report analyzing guidelines proposed by EPA for the Urea and
     Ammonium Nitrate subcategories of the nitrogen fertilzer in-
     dustry are reproduced as Appendix A to the Centec Report
     submitted herewith.

2/   Excerpts from this Jacobs Engineering Report dated  August 18,
     1976 are reproduced as Appendix B to the Centec Report sub-
     mitted herewith.

V   42 Fed. Reg. 16140-16141, a copy of which is reproduced as
     Appendix C to the Centec Report submitted herewith.
                                -8-

-------
               The information presently available  to
          the Agency indicates that while the better
          nitric acid plants are able to maintain their
          effluent discharges within the pH range of
          6.0 to 9.0 more than 99.9 percent of the  time
          on a yearly basis, maintaining continuously
          or 100 percent of the time does not now appear
          to be economically feasible.

          On April 26, 1978, EPA promulgated amended regulations

for the Ammonium Nitrate and Urea Subcategories.  Despite  its

recognition in the March 25, 1977 Notice of Revocation of  the

Nitric Acid and Phosphate fertilizer regulations  that  maintenance

of pH 6.0 to 9.0 on a continuous basis was not "economically feas-

ible", the new regulations contain a requirement  that  pH be  main-

tained "[wjithin the range 6.0 to 9.0" (43 Fed. Reg. 17821-17828),

This requirement was issued even though the Agency's own recently

completed survey demonstrated the inability of fertilizer  plants

having a discharge to meet the pH 6 to 9 limitation on a contin-
           !/
uous basis.

          Although, none of the Agency's development documents

supportive of effluent guidelines and standards containing con-

tinuous pH 6.0 to 9.0 requirements have offered technical

support for the requirement, Agency policy has been clear.

Effluent guidelines have consistently contained a 6.0  to 9.0
I/   Of the 17 plants surveyed which continuously monitored their
     effluent, only four reported no pH excursions.   Three of
     those plants had no discharge of any pollutants and thus,
     by definition, no pH control problem.  The fourth initially
     reported that it was meeting the pH 6 to 9 requirement on
     a continuous basis.  This was inaccurate, however,  and was
     clarified by a letter from the plant to the Agency.  It
     should be noted that this plant has the most advanced tech-
     nology the Agency has identified as available.
                                -9-

-------
continuous pH requirement.  For example, the recently promulgated

fertilizer regulations contain such a requirement.  Likewise,

Agency policy in the absence of guidelines has been clearly

enunciated.  A July 15, 1976 memorandum to "Regional Enforcement

Directors NPDES Approved State" from Jeffrey G. Miller,  Deputy

Assistant Administrator for Water Enforcement re_ "pH Limitation

in NPDES Permits" states:

               Generally, our position has been that for
          BPCTCA, the pH limitations should be 6.0 to
          9.0, unless of course, relevant effluent guide-
          lines limitations for pH are different or else,
          water quality standards are more restrictive.

This policy has been followed in issuance and enforcement of NPDES
        !/
permits.   Its support appears limited to a June, 1976 memorandum

entitled "Justification for pH Limits in NPDES Permits'"  prepared by

Murray P. Strier, an Agency staff chemist.  This memorandum does

not even consider (1) the technical justification for a  6.0 to 9.0

standard, (2) the costs of attempting to achieve that standard,

(3) the necessity for provision for excursions from the  required

standard, or (4) the alleged harm to receiving waters from discharge
I/   For example, in 1974 EPA issued an NPDES permit containing
     a pH 6 to 9 100 percent continuous monitoring requirement
     for the Wurtland, Kentucky,  sulfuric acid plant of E.  I.
     du Pont de Nemours and Co.,  and Region IV threatened to
     rigidly enforce this permit  provision.  On October 25, 1977,
     du Pont filed suit challenging this limitation in the  United
     States Court of Appeals for  the Sixth Circuit.  This liti-
     gation was only recently settled through entry of a stip-
     ulated decree providing inter alia that "(a)  the pH limita-
     tion [of 6.0 to 9.0] shall not be exceeded more frequently
     than 1% of the time the plant was operating during each
     month or 10% of any day (24-hour operating period)."  E.  I.
     du Pont de Nemours & Co. v.  United States Environmental
     Protection Agency, No. 77-3507 (6th Cir., May, 1978).
                              -10-

-------
 of effluent  deviating  from  the  required  standard.  Without any con-

 sideration of  these  issues,  the memorandum simply states:

           .  .  .  Therefore,  it must be concluded that pH
           control  of industrial waste streams are open
           to many  relatively facile and  inexpensive
           options.   (p.  6)

 Though  this  position was recognized by the Agency to be unsound

 in the  case  of nitric  acid  and  phosphate fertilizer plants in the

 March,  1977  Notice of  Revocation, it is apparently still being

 followed.

           The Centec Report  submitted herewith corroborates with

 extensive  operating  data the conclusion of both the June, 1976

 Centec  Report and  the  September, 1976 Jacobs Engineering Report.

 On authority of  these  studies,  maintenance of pH 6.0 to 9.0 on a

 continuous basis is  not  feasible for the specified subcategories.


     C.    The 1977 Amendments

           In the 1977 Amendments to the Act, enacted on December 27,

 1977, Congress significantly revised the regulatory system for the

 control of effluent  discharges  under the Act.  The 1972 Act required

 the adoption of best available  technology (BAT)  for all point sources

 by July 1, 1983.   Congress, however,'was persuaded that this might

 prove unnecessary and too costly for the control of so-called "con-

 ventional pollutants".  Accordingly, the 1977 amendments require

 EPA to develop regulations specifying best conventional pollutant

 control technology (BCT) for conventional pollutants,  including

pH.   The Agency  is to review existing  BAT regulations for conventional
     Section 304(a)(4) of the Act,  as amended.
                              -11-

-------
pollutants and to make any necessary adjustments  to  effectuate

BCT.  For 21 major industrial point source categories,  this

review must be accomplished by July 1,  1980;  for  all other point

source categories, it was to have been  completed  by  March 27,

1978.  BCT regulations for these conventional pollutants must

be complied with by January 1, 1984.

          In developing BCT for classes and categories  of point

sources, the Agency is directed to make a cost-benefit  analysis

as well as to consider other factors, including age  of  equip-

ment, engineering considerations, and process changes.   Regula-

tions more stringent than BPT may be issued only  if  the benefits
                                                  I/
of such regulation outweigh the incremental costs.    The purpose

of this analysis is to limit the control required for conventional

pollutants to that achievable at reasonable cost. As discussed

below, this level of pH control for the types of  plants examined

by Centec is maintenance of pH within the range 6.0  to 9.0  for

99 percent of the time on a monthly basis with allowance  for

excursions of short duration outside the 3.5 to 11.0 range.



III.  GROUNDS SUPPORTING THIS PETITION

          While Petitioners, and industry generally, have com-

piled impressive records of pH control, maintenance  of effluent

discharge within pH 6 to 9 at all times is not feasible for

plants which manufacture strong acids or substantial quantities
I/   Section 304(b)(4)(B).
                              -12-

-------
of bases or utilize these materials in their manufacturing

processes.  The Centec Report presents data reflecting the unsuc-

cessful efforts of several plants in the specified subcategories

to maintain pH within the Agency's parameters on a continuous

basis.  Despite installing highly sophisticated equipment for

pH control, the data indicates failure in each instance to elim-

inate brief excursions from the 6 to 9 range.  All available data

and information indicate, however, that these brief excursions

have no meaningful adverse environmental effect on receiving

waters.

          A summary discussion of these conclusions, all of which

are extensively documented in the Centec Report, is provided below.


     A.   Maintenance Of pH Within The 6-9 Range At All Times
          Is Not Feasible With Presently Available Technology

          The pH scale within the range of 6 to 9 is very sensi-

tive to the addition of small amounts of acid or base in unbuffered

or slightly buffered water.  Maintenance of industrial discharge

within this range on a continuous basis is extremely difficult for

companies which manufacture strong acids or substantial quantities

of bases or which use them in their manufacturing processes.  (Centec

Report §§ 2.1, 2.3.)

          Most natural waterways have a high buffering capability

which effectively maintains pH near the neutral range even when

excess hydrogen or hydroxyl ions are added to it.  Since most

highly acidic wastewater has had its buffering capacity destroyed,

however, this same buffering effect will not be apparent in the
                              -13-

-------
wastewater itself.  This factor accentuates  the  problem  of main-



taining the pH of effluent discharge  within  the  6  to  9 range



on a continuous basis.  (Centec Report §§  2.2,  2.3.)



          Control of pH is essentially a matter  of neutralizing



the wastewater by addition of base (if acidic)  or-acid  (if basic).



Because the makeup of some industrial wastewaters  is  continually



changing, controlling pH requires responding to  dynamic  conditions,



The sensitivity to pH change of wastewater near  the neutral  range



and its generally diminished buffering capability  make  immediate



and precise response to these changing conditions  necessary  to



maintain pH within 6 to 9.  (Centec Report § 2.3.)



          The most advanced techniques and mechanisms developed



to control pH include feedback and feed-forward  systems  and



attenuation tanks or ponds.  These systems,  which  are generally



used in varying combinations determined by a particular  plant's



situation, are discussed in detail in Section 4  of the Centec



Report.  The efforts of 11 plants in the specified subcategories,



representing 18 years of pH control experience, to control pH



through use of various combinations of these systems, is dis-



cussed in Section 5 of the Centec Report.   Though  use of these



systems has produced very favorable results in terms  of reduc-



tion of excess hydrogen or hydroxyl ions in the effluent and



percent of time in compliance within pH 6 to 9, none  of the



plants has been able  to attain 100 percent compliance with pH



6  to 9 requirements.  Even the precise system recommended by



EPA's contractor, which has been  installed at U.S. Steel's
                              -14-

-------
 Pittsburg, California plant, has failed to meet the Agency's


 requirement.  Another of the plants included in the Centec study

 has  installed at great cost a system to divert and recycle efflu-

 ent  emitted from the system that is outside the pH 6 to 9 range.

 Even with the addition of this system, maintenance of the pH


 6  to 9 requirement on a continuous basis has not been achieved.


          The reasons for the inability of many plants to main-


 tain effluent discharge within pH 6 to 9 on a continuous basis


 are discussed in Section 4 of the Centec Report.  The problems


 involved include effluent variability, variability of titration


 curves, electrode fouling, unstable control, mechanical failure,

 and operator error.  These factors make 100 percent reliability


 of any pH control system impossible to achieve.  Without 100

 percent reliability, maintenance of pH within 6 to 9 on a con-


 tinuous basis is unattainable by plants in the specified sub-

 categories .

                                        !/
          Table 6-3 of the Centec Report   summarizes the per-


 formance and cost of advanced pH control for various plants.  The

 installed cost of these systems ranges from $60,000 for a single-

 stage automatic control system to over $4  million (1977 dollars)

 for a multi-stage system with recycling.  Many of the plants


achieved outstanding pH control levels, often better than 99 per-

cent on a long-term (yearly)  basis.   Clearly these plants are
I/   Centec Report, p. 156.
                              -15-

-------
operating with an extremely high degree  of  reliability with little
room for improvement in pH control.   As  would  be  expected, given
the sporadic nature of the factors contributing to pH excursions,
monthly averages for time in compliance  within the pH 6  to 9  range
were at times lower than the yearly averages.   Only  two  of the
plants for which monthly compliance data was available consist-
ently achieved better than 99 percent compliance  on  a monthly
basis.  Installed costs of these systems were  $531,000 and $800,
000, respectively.
          Generally, the available data  indicate  that costs
rise rapidly as plants attempt to improve pH control past 98
percent of the time on a monthly average basis.  In  the  opinion
of Centec, 99 percent compliance with a  pH 6 to 9 requirement on a
monthly basis can be met by most plants  in the specified subcat-
egories with one or two stages of neutralization, combined with
attentuation tanks or ponds.  Such a level of compliance thus
represents the upper limit of effluent reduction achievable by both
BPT and BCT.  The very substantial capital, land, technical manpower
and energy resources required for diversion ponds, redundant systems
and other methods of improving  control beyond this point would be
wasted  since  a requirement of 99 percent compliance will adequately
protect the environment.   (Centec Report, p.  7.)
                               -16-

-------
      B.    The Environmental Benefits Of pH Control Beyond
           99 Percent On A Monthly Basis Are Non-Existent

           1.    Short-Term pH Excursions from the 6.0
                to  9.0 Range Typical of Even the Best
                Controlled Plants Will Not Have A
                Harmful Effect on the Environment	

           Though very high or low concentrations of hydrogen

 ions  in water can  be harmful to aquatic life under certain con-

 ditions, the effect is determined by the magnitude and duration

 of  the exposure.   A survey of the available scientific literature

 indicates  that  the limits of mortality for various species of fish

 extend to  pH 4  for several hours.  This is a much longer time at

 this  pH level than usually experienced in industrial situations.

 Furthermore, pH is not bioaccumulative, i.e. , short-term exposures

 are not cumulatively harmful.  This is in contrast to the long-

 lasting effects of toxic substances, such as lead or mercury,

 where small amounts accumulate in the food chain.  An extensive

 literature search reveals no data showing any permanent effect

 attributable to brief exposures to low pH.  The effect of pH above

 9 is  not as well documented, but it appears that even continuous

 exposure of pH up to 9.8 is not damaging.   (Centec Report § 3.)

          The ability of aquatic life to accommodate short-term

 extreme pH conditions in a stream and the ability of fish to

 avoid low pH areas which affect only a portion of the stream

 support the judgment that aquatic life would not be adversely

 affected by short-term (up to several hours) excursions from the

pH range 6 to 9.  Available surveys of well-controlled industrial
                              -17-

-------
effluents that experience short-term  excursions  from pH 6 to 9

show no discernible effect on the  environment  as a result

thereof.  (Centec Report § 3.)


          2.   Controlled Industrial  Discharges
               Represent Inconsequential  Sources
               of pH Change in Receiving  Waters

          Buffering is the phenomenon by  which compounds dis-

solved in water mitigate the effect on pH of added hydrogen

or hydroxyl ions.  In unbuffered water the addition of an acid

or base results in a proportional  increase or  decrease of the

free hydrogen ions.  In buffered water, however, the dissolved

compounds already present in the water react in  such a way that

the pH change is much less than would be  the case in an unbuf-

fered solution.  (Centec Report §  2.2.)

          Because of buffering and rapid  dilution, the addition

of small quantities of excess hydrogen or hydroxyl ions to natural

waterways has only a minimal effect on the pH  of the receiving

waters.  The effect that is apparent  is localized in nature  and

of short duration.

          Section 7 of the Centec  Report  documents effects on

receiving water pH of industrial discharges outside  the 6  to 9

range.  Each case studied involved excursions  from the  6  to  9

range much more egregious than typical industrial discharges

from a plant with advanced pH control.  Nevertheless,  in  each

instance effect on pH of the receiving waters  was very localized

and of short duration.  No environmental  harm  was evident.
                              -18-

-------
     C.    EPA's pH Regulations and  pH  Policy
          Are Not Consistent With the  Act

          Centec's analysis of the  most  advanced methods of

pH control and of the actual performance of plants  that have

installed them leads to the conclusion that no control system

exists that is capable of the essentially  perfect control  that

is necessary to maintain a continuously varying industrial dis-
                                               l/
charge within the sensitive pH range of  6  to  9.  Because  of

inherent limitations of pH control  technology, plant upsets,

mechanical breakdowns and other factors beyond the  control of

plant operators, pH cannot be maintained within the 6 to 9 range

100 percent of the time.  A failure by the Agency to make  allowance

for limited excursions from this range in  existing  and new pH

regulations for the designated subcategories  would  be unrealistic

and improper.  See, e.g., Marathon Oil Co. v. EPA,  564 F.2d  1253,

1272 (9th Cir. 1977); Tanners Council  of America v. Train, 540  F.2d

1188, 1194 (4th Cir. 1976); FMC Corporation  v. Train, 539  F.2d  973,

986 (4th Cir. 1976); Essex Chemical Corporation v.  Ruckelshaus,

486 F.2d 427, 432 (D.C. Cir. 1973), cert,  denied, sub nom

Applachian Power Co., et al. v. Environmental Protection Agency,

416 U.S. 969 (1974); Portland Cement Ass'n v. Ruckelshaus, 486  F.2d

375 (D.C. Cir. 1973), cert, denied, 417 U.S.  921 (1974).   In the

Marathon Oil case, the court was reviewing effluent regulations

applicable to offshore oil drilling platforms.  The court  found

that while ,the standards could be met  from 97.5 - 99 percent of


I/   Centec Report, p. 6.
                              -19-

-------
the time they could not be met 100 percent of the time because

of various upsets.  The court stated:

          .  . .  The issue raised by petitioners is whether
          a  formal "upset provision" must be written into
          the permits.  The EPA has refused to insert an
          "upset provision" into petitioners' permits,
          arguing that excursions can be adequately dealt
          with informally.  Petitioners argue that this
          is not enough:  that the permits must formally
          provide that upsets beyond the control of the
          permit holder are not violations of the permit
          standards.
               We agree.  The Federal Water Pollution
          Control Act requires point sources of pollution
          to utilize  the "best practicable control
          technology  currently available" prior to 1983.
          The EPA cannot impose a higher standard without
          violating the Control Act.  And yet the permits
          as currently written do exactly that.  The
          permits on  their face require petitioners
          to meet the standards 100 percent of the time.
          But platforms using BPCTCA can only be expected
          to achieve  the effluent standards 97.5 percent
          of the time in the case of deck drainage and 99
          percent of  the time in the case of produced water.
          We, therefore, remand to the EPA with instructions
          to insert upset provisions into the permits.
               It is  not an adequate response that the EPA
          will informally take BPCTCA into account in deciding
          whether or  not to prosecute "violators."  First,
          there  is no guarantee that the EPA will choose
          to exercise this discretion.  And once a prose-
          cution is brought, the courts have no authority
          to dismiss  the complaint on the grounds that the
          permit holder could not have avoided the violation.
          Cf. Getty Oil Co. v. Ruckelshaus, 467 F.2d 349  (3d
          Cir. 1972), cert, denied, 409 U.S. 1125, 93 S.Ct.
          937, 35 L.Ed.2d 256 (1973).   Even if we were to
          assume that the EPA would decline to prosecute in
          every case  of unavoidable excursion, any concerned
          citizen would still be free to commence an enforce-
          ment action against the "violator" under section
          505(a)(l) of the Act.  "[T]here is no authorization
          to block a  citizen's suit under section 505 even
          though the  agency believes that the suit should
          not go forward."  Bethlehem Steel Corp. v. Train,
          544 F.2d 657, 660 (3d Cir. 1976).
          (564 F.2d at 1272-1273, footnote omitted.)
                              -20-

-------
          It is important to note that unlike the situation  in



the Marathon Oil case, petitioners are not merely seeking  an



amendment to the Agency's pH regulations to cover plant upsets



and mechanical breakdowns.  In the 1977 Amendments to the  Act,



Congress specifically designated pH as a pollutant parameter



which should be subject to BCT rather than BAT.  It is therefore



essential to determine what level of pH control can be achieved



consistently at a reasonable cost.  Under the Act, both BPT and



BCT regulations must reflect a level of performance that can



be economically and consistently achieved.  See, E. I. du Pont



de Nemours & Co. v. Train, 541 F.2d 1018, 1035 (4th Cir. 1976)



(remanding BPT and BAT pH regulations), decision on other issues



rev'd, 430 U.S. 112 (1977).  In numerous regulations, the Agency



has recognized that effluent guidelines must be computed by



multiplying a long-range average performance figure by a



variability factor (e.g., 43 Fed. Reg. 17821 (April 26, 1978)



(urea and ammonium nitrate plant regulations)).  The Centec



Report demonstrates that pH control technology has inherent



limitations when applied to certain types of industrial effluent



which result in variations in performance.   It is therefore



necessary to take such variability  into account  in deriving pH



regulations.  For unstated reasons, however, EPA has not used



that procedure to derive  its pH regulations.



          While an appropriate pH regulation can be  set for a



given industrial category or subcategory only  after  examining



the data from a representative number  of plants, the Centec
                              -21-

-------
Report indicates that 99 percent compliance on a monthly basis


is the maximum performance that can be economically achieved


by most plants.  Only a few of the plants studied by Centec were


achieving that level of performance consistently.  Some plants


not specifically reviewed by Centec may not be able to achieve that

level consistently because of the volume of their discharge, avail-

ability of land for attenuation ponds, hydraulic profile or other


factors.


          Furthermore, all plants will have difficulties in


achieving even a 99 percent level of control during periods in


which there are severe plant upsets or mechanical breakdowns.

pH control systems are only designed to deal with normal variations


in influent and are themselves subject to mechanical breakdowns

(Centec Report § 4.3).  An upset provision therefore is required


for those occasions when even exemplary use of BPT or BCT will not


avoid an excursion.  Marathon Oil Co., supra.   In most cases,.


however, excursions outside the pH range 3.5 to 11 caused by plant


upsets should be able to be brought under control within 15 minutes,
                               !/
if necessary by plant shutdown.   An upset provision by itself

would thus provide only limited relief from the current 100 percent

compliance requirement.


          Because the current regulations and  policy require a


level of performance that cannot be consistently achieved and
I/   Some plants may require longer than 15 minutes to shut down.
     In such cases, it would be unreasonable to limit such
     excursions to a shorter period than that required to shutdown.
                              -22-

-------
fail to make any provisions for plant upsets,  they are  not




authorized by the Act.








IV.  RELIEF REQUESTED BY THIS PETITION FOR REVIEW AND REVISION




          This petition requests the Agency to (-1) revise its




internal policy and existing regulations and (2)  issue  new  guide-



lines where none are currently in effect to allow industrial



dischargers in point source categories involving  the manufacture,




or use in the manufacturing process, of strong acids or substan-




tial quantities of bases some modest relief from  the Agency's




requirement that pH be maintained within the 6 to 9 range on




a continuous basis.



          The Centec Report establishes that for  most plants



studied it would be reasonable to require maintenance of pH 6



to 9 for 99 percent of the time on a monthly basis provided that




excursions outside pH 3.5 to 11 did not exceed 15 minutes duration.



The extensive data presented indicates that implementation of this



recommendation would have no negative environmental impact and that



these parameters are generally attainable although at considerable



cost.  Though only two of the plants studied by Centec are currently



meeting the pH requirements suggested, the Centec Report states



that most plants in the subcategories studied should be able to




do so through the use of the advanced pH control  technology identi-




fied in the Report.  It must be emphasized, however, that the
                              -23-

-------
recommended pH parameters represent the maximum pH control attain-
able at an economically feasible cost.  They do not take into
consideration the occasional severe plant upset or equipment
malfunction which may make it impossible to meet a 99 percent
compliance level.  The new regulations therefore must also contain
a limited upset provision covering such situations.
          The revision of Agency policy hereby requested requires
the withdrawal by the Agency of the July 15, 1976 memorandum from
Jeffrey Miller to Regional Enforcement Directors r_e "pH Limita-
tions in NPDES Permits", discussed above, or the issuance of a
memorandum clarifying its applicability.
          Since effluent limitations guidelines and standards of
performance for many of the point source subcategories addressed
in this petition are currently under remand to the Agency, new
pH guidelines for these subcategories should be issued which
contain pH parameters consistent with the recommendations contained
herein.
          Where regulations are now effective for the specified
subcategories, the pH parameters contained therein should be
revised.  These existing regulations'are subject to judicial
review under § 509(b) of the Act on the basis of the new data
and information contained in the Centec Report that is submitted
herewith.   However, to avoid unnecessary litigation, we are first
submitting this information to the Agency and requesting that
appropriate modifications in those regulations be made.
                               -24-

-------
          It is imperative that these Agency actions be accom-

plished promptly in order to provide permit-issuing authorities

necessary guidance in issuance of NPDES permits containing reason-

ably achievable pH parameters.

                                Respectfully submitted,

                                CLEARY, GOTTLIEB, STEEN & HAMILTON

                                Attorneys for Petitioners
OF COUNSEL:

  Richard deC. Hinds
  Douglas E. Kliever
  John S. Magney
  1250 Connecticut Avenue, N.W.
  Washington, B.C.  20036
  (202) 223-2151

Dated:  August 3, 1978
                              -25-

-------
                     pH  UNDER  CONTINUOUS MONITORING

Background

Categorical  standards for  effluent guidelines generally require that the
pH  of plant  effluent be maintained in the range of 6 to 9, and some
permits  require  that pH be monitored continuously.  Although data indicate
that achievement of  this standard all of the time is not possible, a.
standard can be  established for  a percentage of time over which a plant
                                                             *     *
must achieve the required  range.

The findings of  this report support two standards with regard to continuous
monitoring of wastewater pH:

     (1)  Plants with required continuous monitoring are not to be
          outside of the categorical pH range more than 432 minutes,
          or one per cent, of each month.

     (2)  All continuous periods during which the final effluent
          is outside the categorical range that exceed 15 minutes
          in duration are  to be reported to the permit authority.

The lack of  perfect  compliance with the standard of pH 6-9 at all times.
is in large  part due to the existence of continuous monitoring for pH.
Since perfect control of many pollutants is impossible, it has been
Agency policy to  set a  standard for which the probability of violation

-------
at a given Instant by a:we11 designed and maintained plant is  not greater
than .01.  When the monitoring schedule specifies grab or composite
samples, the  detection a departure from the standard is a rare event
because the departure must coincide with the time of sampling.   With
continuous monitoring, however, all departures will  be detected, and
consequently, known deviations outside the categorical range are not
rare events, although the plant may be within the range almost all of
the time.  Categorical pH standards, therefore, are  consistent with
other effluent guideline standards and Agency policy; however, the
introduction of continuous monitoring into permit requirements necessitates
a clarification of policy to determine appropriately whether a_plant  is
in compliance.

To secure data on the performance of pH control in several industries,
the Agency collected continuous monitoring data from six plants.  Each
of these plants provided one year of strip charts from the pH recorders.
Data were also collected on the types of treatment performed at each
plant, the reason for each excursion (defined below) from the range 6-9,
and the costs of pH control systems.  Results of this data collection
are documented in the draft contractor's report, An  Assessment of pH-
Control of Process Waters in Selected Plants, by JRB Associates, March
1979.

An excursion is a continuous period when the plant's effluent is outside
the range 6-9.  For each excursion, the date, the time the excursion
began, the reason code (See Table 1), and the total   time  in excursion
were entered into a data base.  This data base constitutes the  information

-------
used in this report for .'evaluation of pH standards coupled with continuous
monitoring.

Analysis of Monthly Compliance Time.

A requirement that pH 6-9 be maintained 99 per cent of the time is
consistent with existing Agency policy for setting effluent guidelines.
According to Table 2, each plant in the data base meets the standard at
least 99 per cent of the time on an annual basis.  However, there are
plants in the data base which did not achieve 99 per cent compliance in
every month; the following analysis addresses the question of variability
on  a month-to-month basis.

An  examination of the empirical distribution functions of lengths of
individual  excursions reveals that  two  types of  distributions  exist,
depending  on the reason code.   Excursions of reasons codes 1,  2, 4, and
5 were found to have an approximately exponential distribution with a
mean  to  be  determined from the  data for each plant.  Excursions with
reason code 6 were  found  to  have  positive probability  of  being one minute
(the  limit of resolution  for reading from strip  charts) or less, with
the time  in excess  of one minute  having an  approximately  exponential
distribution.   (Reason  Code  3  is  excluded from this  analysis  because
"excursions" with this  reason  code  have no  flow  associated with them.)
It  was assumed  that the number of excursions  from reason  codes 1, 2, 4
and 5 is  a  Poisson  random variable  with a mean to be determined from the
data  and  that the  number  of  excursions  from reason  code 6 is  Poisson
with  not  necessarily  the  same mean. These  assumptions reflect the  fact
that  the  total  monthly  time  in excursion  is a  sum of random  number  of
individual  excursion  times.

-------
Because the distribution function of total  excursion time does  not  have
a simple, closed-fora mathematical solution, a sinulation of 1000
months of pH excursions with reason codes 1, 2, 4, 5 and 6 v/as  run  for
each plant.  From each sinulation, the 99th percentile for the  total
monthly excursion times was selected.  As seen in Table 3, three
of the six plants achieved less than 432 minutes per month (This is one
per cent of a 30 day month.)  in excursion; the median across plants of
these 99th percentiles is 379 minutes per month in excursion.  Where
previous effluent guidelines have been based on the performance at  several
plants, the median across plants has generally been used as a basis for
a standard.  Following this method, a standard which allows onei per cent
of a month, or 432 minutes, in excursion is approximately the same  as
the median of 379 minutes of plant-specific 99th percentiles.  ( It is
also to.be noted that the analysis performed by Centec Consultants  in
response to the ORB report is also consistent with a policy allowing
excursions one per cent of the time.)

In computing the mean lengths of excursions, three data points were
deleted as outliers, because they were associated with very long excursions
.(relative to other excursions at the same plant) which would not be
typical of regular plant operations.  Specifically, an excursion of 1900
minutes was deleted fron the data for Plant 8011; and excursions of 720
and 330 minutes were deleted from the data  for  Plant 3141.   The simulations
were run using mean excursion lengths computed  without outliers.

-------
The implications of a requirement that excursions exceeding 15 minutes in
length are presented here and in Tables 5, 6, and 7.  Table 5 presents
aggregate statistics on the numbers of excursions by Reason Code and the
percentage within each code which exceed 15 minutes in length.  These range
from 71 per cent of the relatively rare excursions due to process upset to
less than 3 per cent of the relatively frequent excursions due "to instrument
error.  Summing the expected numbers of excursions in the last column, the
requirement would entail, in aggregate, approximately 14 reportable events
per plant per year.  More detail is provided in Table 6, which deals only
with excursions in Reason Codes  1, 2, 4, 5, in which excursions.in excess of
15 minutes are relatively more frequent.

Table  7  includes  excursions  from all Reason Codes  (except 3)  and reflects the
consequences  of the  15-minute  reporting requirement.  The reason code which
contributes  the greatest number  of  excursions  in excess  of 15 minutes is 5,
the code  for "unknown"  reasons.  All but  a few of  the excursions are  for reasons
related  to  the malfunction  of processes or treatment systems.

-------
                               Table  1.

                         Reason Coding  System
Reason Code                                Reason Type
   1                                 Process Upset
   2                                 Treatment System Upset
   3                                 Technical Excursion-Excursion  recorded,
                                     but flow recording charts  indicate no flow
                                     to  surface water
   4                                 Other
   5                                 Unknown
   6                                 Instrument Error

Note:  The word upset, as used here,  refers to any malfunction of a  manufacturing
process or treatment system and is not  to'be-confused with occurrences defined
in "Upset and Bypass" regulations.

-------
                                 Table 2.
                     percent Annual Compliance Times
                         for Selected Data Subsets
Plane
1305
2128
i
2653
3141
6662
son
Average
Percent Annual
Compliance Tine —
All Reason Codes
99.8
100.0 (1)
99.4
99.4
99.7
99.2
99.6.
Percent Annual
Compliance Tiae —
Codes 1, 2, 4, 5
100.0 (1)
100.0- (1)
99.4
99.5
99.7
99.6
99.7
Percent Arr^si
Compliance Tim-e
Codes 1, 2
100.0 (1)
100.0
99.5
99.5
100.0. (1)
99.7
99. S
(1)   Calculated percent vas 99.95 or larger
                Reason
                 Code
                   2
                   3
                   4
                   5
                   6
    Season
Process l^pset
Treatment System Upser
Technical Excursion
Other
Unknown
Instrument Error or Calibration

-------
Table 3.  Montly Frequencies and Average Lengths of Excursions With
          Simulated 99th Percentile of Monthly Time in Excursion.
            Monthly Frequency of Excursion     Mean Length of Excursion
Plant
ID
1306
2128
2653
3141 (2)
6662
8011 (3)
Average
Median
Reason Codes
1, 2, 4, 5
0.50
0.17
1.75
11.27
1.08
1.0
2.63
1.04
Reason Code
6
22.75
0
10.58
1.5
0
4.5
6.57
3.00
Reason. Codes
1, 2, 4, 5
22.0
24.5
136.2
17.7
115.4
22.3
56.4
23.4
All Reason Code
6
1.72
0(1)
1.06
16.17
0(1)
12.67
7.91
7.20
Reason Code 6 and
greater than 1 min
5.48
. 0(1)
3.33
19.2
0(1)
53.5
20.38
12.34
Minute

143
62
1200
463
782
316
494
379
Notes:

(1)  No excursions reported in this category.
(2)  Two longest excursions deleted from the  data  base  as  outliers.
(3)  Longest excursion deleted from the data  base  as  an outlier.
Simulated 99th
Percentile,

-------
Table 4.  Numbers of Excursions by Plant,  Reason Code, and Length


                    Number of Excursions
Plant
ID
1306
2128
2653
3141
6662
8011
Reason Codes
1. 2, 4 5
6
2
21
134(1)
13
12(2)
Reason Code
6
273
0
127
18
0
54
Reason Code 6:
Longer than 1 nrin.
44
0
3
15
0
12
 Notes:

 (1)   Two outliers  deleted.
 (2)   One outlier deleted.

-------
                             Table 5

                        Aggregate  Excursion Statistics
Reason Code
1.
2.
3.
4.
5.
6.
7.
Process
Treatment
Technical
Other
Unknown
Instrument
All
Number Percent Exceeding 15 Min.
17
44
40
8
122
472
703
715
43%
45%
705
26%
2.8%
12.5%
Expected Number Greater T
15 Minutes Per Plant Per
2.0
3.2
3.0
1.0
5.3
2.2
13.8
Technical excursions could be  avo.ided by  installation of pH  r-ecorders
dedicated to monitoring.   The  level  of  15 minutes - plus excursions due
to instrument error is small,  and  the other could be subject to upset
and'bypass provisions given necessary action and'justification.

-------
                                    Table 6
Distribution of Excursions  in  Reason Codes 1, 2, 4, and 5 by Plant
                               Number of Excursions
Plant ID     Total        Greater  than 15 Hin.     Percentage Exceeding 15 Hin.
1306           63                         50
2128           2                 1                         50
2653           21                18                         86
3141          136                33                         24
6662           13                9                         69
8011           13                6                         46

-------
Table 7.  Distribution of-Excursions Exceeding 15 Minutes by Plant
and Reason Code.
           Reason
             Code
Plant
1306
2128
2653
3141
6662
8011
TOTAL
                                                     TOL
2
0
10
-0
0
0
1
0
3
4
9
2
0
1
2
0
0
4
0
0
3
29
0
0
4
0
0
5
0
4
7
1
18
38
9
10
12
19
32
13
83

-------
     Table  8.
CUMUUVTIVS  FREEUENCY I'l STRI S.UTION
        FOR  REASON ccr-e  :
              <1 2 4  5>
T=T.
'." 0
1. 0
2. 0
3. 0
5.0
6.0
7.0
"*—* ri
10. 0
12. 0
13.0
14. 0
15. 0
20. 0
21. 0
22. 0
25. 0
3 0 . 0
35. 0
36. 0
37. 0
40. 0
43. 0
50. 0
60. 0
65.0
70.0
73.0
SO.O
90.0
95. 0
1 05. 0
120. 0
125.0
140.0
142.0
"160. 0
195.0
204. 0
208. 0
210.0
240.0
266.0
"~2"93.' 0
325.0
330. 0
420. 0
720. 0
1900.0
CUMULATIVE
FREQUENCY
191. 0
188. 0
180. 0
156.0
127. 0
126. 0
113. 0
110. 0
92.0
90.0
87. 0
86. 0
70. 0
62. 0
60. 0
57.0
54. 0
46.0
44.0
43. 0
42. 0
40.0
39. 0
33. 0
36. 0
35.0
32.0
30. 0
29.0
27.0
25.0
23.0
20.0'
19VO
13.0
17.0
16.0
15.8
14. £
13. -0
ll.O
7."0
6^0
"S.'O
-Co
3.0
2.0
!-•$
.0
PP.aE.riZ.IL ITY
1. 000
.984
.942
.817
. 665
.660
. 592
.•576
.482
.471
.435
.430
. 366
. 325
.314
. 298
. 283
.241
.230
.225
.220
.209
. 204
.199
. 133
. 183
. 163
.157
. 152
. 141
. 131
.120
. 1 05-
".1399
. 094
. 089
.Oc4
. 079
.073
. 068
. 058
.037
-031
.026
.021
.016
.010
-003
OOD

-------
                TABLE 9
     CUMUI_i=i7I'/E  FREQUENCY  3D I 3TR I BUT I DM
               FDR  REAs-ON  COI'E fX"f< :
 IiijRftTIDN        CUMUt-ATI'/E        F'RD2:ftZ-11- I TY
   •;' M I N '••           F RE E U E N C Y
    T=TO            N<;T>TO>               £T;.TB;I















1
1
•
1 .
C. •
C. »
.— ,
'•' *
^
*t •
5.
i* •
*— * •
10.
1 3 .
20.
3 0 .
•^••— • •
35.
40.
35.
* 0 .
3*0.
ij
0
0
5.
0
0
o
0
0
0
0
0
0
0
0
0
0
0
0
472.
74.
44.
4
.*,
•••'
.£
*•* •
4.
1.
24.
c
1
1
1
1







0.
y »
^t
•»' •
.»,
c, *
0.
7 .
6.
5.
•"•
C* »
'— *
L_ •
1.
•
o
0
0
0
0
0
0
0
0
0
o
o
0
o
o
0
o
0
0
1 . Ci 0 0
.157
. 093
. 091
. 072
. Ot-6
. 051
. 042
. 033
. 023
. 025
. 021
. 015
. 0 1 3
. Oil
.00*
.004
. 002
. 000
1JU.S. GOVERNMENT PRINTING OFFICE: ,1980-341-082/131

-------