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
------- |