vvEPA
United States
Environmental Protection
Agency
Solid Waste And
Emergency Response/
Research and Development
(OS-420) WF
EPA/530/US 1-aO/OUa
March 1990
Standard Test Procedures
for Evaluating Leak
Detection Methods
Liquid-Phase Out-of-Tank
Product Detectors
Printed on Recycled Paper
-------
-------
Standard Test Procedures for
Evaluating Leak Detection Methods:
Liquid-Phase Out-of-Tank
Product Detectors
Final Report
U.S. Environmental Protection Agency
Office of Research and Development
March 1990
-------
-------
This report has been funded wholly or in part by the Environmental Protection
Agency under Contract No. 68-03-3409 to Radian Corporation. It has been
subject to the Agency's review, and it has been approved for publication as an
EPA document. Mention of trade names or commercial products does not
constitute endorsement or recommendation for use.
-------
-------
FOREWORD
HOW TO DEMONSTRATE THAT LEAK DETECTION METHODS MEET ERA'S PERFORMANCE
STANDARDS
The Environmental Protection Agency's (EPA's) regulations for underground
storage tanks require owners and operators to check for leaks on a routine
basis using one of a number of detection methods (40 CFR Part 280, Subpart D).
In order to ensure the effectiveness of these methods, EPA set minimum
performance standards for equipment used to comply with the regulations. For
example, after December 22, 1990, all tank tightness testing methods must be
capable of detecting a 0.10 gallon per hour leak rate with a probability of
detection of at least 95% and a probability of false alarm of no more than 5%.
It is up to tank owners and operators to select a method of leak detection
that has been shown to.meet the relevant performance standard.
Deciding whether a method meets the standards has not been easy, however.
Until recently, manufacturers of leak detection methods have tested their
equipment using a wide variety of approaches, some more rigorous than others.
Tank owners and operators have been generally unzible to sort through the
conflicting sales claims that are made based on the results of these
evaluations. To help protect consumers, some state agencies have developed
mechanisms for approving leak detection methods. These approval procedures
vary from state to state, making it theirunethod nationwide. The purpose of-
this policy is to describe the ways that owners and operators can check that
the leak detection equipment or service they purchase meets the federal
regulatory requirements. States may have additional requirements for
approving the use of leak detection methods.
EPA will not test,.certify, or approve specific brands of commercial leak
detection equipment. The large number of commercially available leak
detection methods makes it impossible for the Agency to test all the equipment
or to review all the performance claims. Instead, the Agency is describing
how equipment should be tested to prove that it meets the standards.
Conducting this testing is left up to equipment manufacturers in conjunction
with third-party testing organizations. The manufacturers will then provide a
iii
-------
copy of the report showing that the method meets EPA's performance standards.
This information should be provided to customers or regulators as requested.
Tank owners and operators should keep the evaluation results on file to
satisfy EPA's record keeping requirements.
EPA recognizes three distinct ways to prove that a particular brand of
leak detection equipment meets the federal performance standards:
1. Evaluate the method using EPA's standard test procedures for leak
detection equipment;
2. Evaluate the method using a national voluntary consensus code or
standard developed by an nationally recognized association or
independent third-party testing laboratory; or,
3. Evaluate the method using a procedure deemed equivalent to an EPA
procedure by a nationally recognized association or independent
third-party testing laboratory.
The manufacturer of the leak detection method should prove that the method
meets the regulatory performance standards using one of these three
approaches. For regulatory enforcement purposes, each of the approaches is
equally satisfactory. The following sections describe the ways to prove
performance in more detail.
EPA Standard Test Procedures
EPA has developed a series of standard test procedures that cover most of
the methods commonly used for underground storage tank leak detection. These
include:
1. "Standard Test Procedures for Evaluating Leak Detection Methods:
Volumetric Tank Tightness Testing Methods"
2. "Standard Test Procedures for Evaluating Leak Detection Methods:
Nonvolumetric Tank Tightness Testing Methods"
iv
-------
3. "Standard Test. Procedures for Evaluating Leak Detection Methods:
Automatic Tank Gauging Systems"
4. "Standard Test Procedures for Evaluating Leak Detection Methods"
Statistical Inventory Reconciliation Methods"
5. "Standard Test Procedures for Evaluating Leak Detection Methods:
Vapor-Phase Out-of-Tank Product Detectors"
6. "Standard Test Procedures for Evaluating Leak Detection Methods:
Liquid-Phase Out-of-Tank Product Detectors"
7. "Standard Test Procedures for Evaluating Leak Detection Methods:
Pipeline Leak Detection Systems" • '- '
Each test procedure provides an explanation of how to conduct the test, how to
perform the required calculations, and how to report the results. The results
s
from each standard test procedure provide the information needed by tank
owners and operators to determine if the method meets the regulatory
requirements.
The EPA standard test procedures may be conducted directly by equipment
manufacturers or may be conducted by an independent third party under contract
to the manufacturer. However, both state agencies "and tank owners typically
prefer that the evaluation be carried out by an independent third-party in
order to prove compliance with the regulations. Independent third-parties may
include consulting firms, test laboratories, not-for-profit research
organizations, or educational institutions with no organizational conflict of
interest. In general, EPA believes that evaluations are more likely to/be
fair and objective the greater the independence of the evaluating
organization.
-------
Rational Consensus Code or Standard
A second way'for a manufacturer to prove the performance of leak
detection equipment is to evaluate the system following a national voluntary
consensus code or standard developed by a nationally recognized association
(e.g., ASTH, ASME, ANSI, etc.). Throughout the technical regulations for
underground storage tanks, EPA has relied on national voluntary consensus
codes to help tank owners decide which brands of equipment are acceptable.
Although no such code presently exists for evaluating leak detection
equipment, one is under consideration by the ASTM D-34 subcommittee. The
Agency will accept the results of evaluations conducted following this or
similar codes as ,soon as they have been adopted. Guidelines for developing
these'standards may be found in the U.S. Department of Commerce "Procedures
for the Development of Voluntary Product Standards" (FR, Vol. 51, No. 118,
June 20, 1986) and OMB Circular No. A-119.
Alternative Test Procedures Deemed Equivalent to EPA's
In some cases, a specific leak detection method may not be adequately
covered by EPA standard test procedures or a national voluntary consensus
code, or the manufacturer may have access to data that makes it easier to
evaluate the system another way. Manufacturers who wish to have their
equipment tested according to a different plan (or who have already done so)
must have that plan developed or reviewed by a nationally recognized
association or independent third-party testing laboratory (e.g. Factory
Mutual, National Sanitation Foundation, Underwriters Laboratory, etc.). The
results should include an accreditation by the association or laboratory that
the conditions under which the test was conducted were at least as rigorous as
the EPA standard test procedure. In general this will require the following:
1. The evaluation tests the system both under the no-leak condition and
an induced-leak condition with an induced leak rate as close as
possible to (or smaller than) the performance standard. In the case
of tank testing, for example, this will mean testing under both 0.0
gallon per hour and 0.10 gallon per hour leak rates. In the case of
vi
-------
ground-water monitoring, this will mean testing with 0.0 and 0.125
inch of free product.
2. The evaluation should test the system under at least as many
different environmental conditions as the corresponding EPA test
procedure. ,
''/.•'
3. The conditions under which the system is evaluated should be at
least as rigorous as the conditions specified in the corresponding
EPA test procedure. For example, in the case of volumetric tank
tightness testing, the test should include a temperature difference
between the delivered product and that already present in the tank,
as well as the deformation caused by filling the tank prior to
testing.
4. The evaluation results must contain the same information and should
be reported following the same general format as the EPA standard
results sheet. •..>.'.
5. The evaluation of the leak detection method must include physical
testing of a full-sized version of the leak detection equipment, and
a full disclosure must be made of the experimental conditions under
which (1) the evaluation was performed, and (2) the method was
recommended for use. An evaluation based solely on theory or
calculation is not sufficient.
vii
-------
ACKNOWLEDGMENTS
This document was written by Ronald D. Achord for the U.S. Environmental
Protection Agency's Office of Research and Development (EPA/ORD) under
Contract No. 68-03-3409. The Program• Manager was Dorothy A. Stewart, the
Project Director was A. Gwen Eklund, and Julia M. Nault was the Work
Assignment Manager. Philip B. Durgin, PhD was the EPA/ORD Project Officer.
VTI1
-------
CONTENTS
Foreword. i i i
/ ,.
Acknowledgments vl i i
X 0000 Standard Test Method for ACCURACY AND RESPONSE TIME FOR LIQUID-
PHASE OUT-OF-TANK PETROLEUM DETECTORS
X 0002 Standard Test Method for SPECIFICITY FOR LIQUID-PHASE OUT-OF-TANK
PETROLEUM DETECTORS
X 0005 Standard Test Method for LOWER DETECTION LIMIT FOR LIQUID-PHASE OUT,
OF-TANK PETROLEUM DETECTORS
X 0004 Standard Practice for PREPARATION OF SYNTHETIC GASOLINE FOR TESTING
OUT-OF-TANK PETROLEUM DETECTORS
IX
-------
-------
Designation: X 0000
Standard Test Method for
ACCURACY AND RESPONSE TIME FOR LIQUID-PHASE OUT-QF-TANK PETROLEUM DETECTORS
1. Scope
1.1 This test method covers determination of accuracy and response
time of liquid-phase out-of-tank petroleum hydrocarbon leak detectors that
utilize ground water monitoring wells.
1.2 This method is applicable to only the components associated with
detection of liquid-phase petroleum releases for detection systems utilizing
multiple operating principles.
1.3 This standard may involve hazardous materials, operations, and
equipment. This standard does not purport to address all of the safety
problems associated with its use. It is the responsibility of the user of
this standard to establish appropriate safety and health practices and
determine the applicability of regulatory limitations prior to use.
2. Referenced Documents
2.1 , ASTH Standards:
-------
D 1125 Standard Test Methods for Electrical Conductivity and
Resistivity of Water
E 1 Standard Specification for ASTM Thermometers
E 456 Standard Terminology Relating to Statistics
2.2 EPA Leak Detector Standards:
X 0004 Standard Practice for Preparation of Synthetic Gasoline
for Testing Out-of-Tank Petroleum Detectors
3. Terminology
3.1 Definitions — For formal definitions of statistical terms, see
Terminology E 456.
3.2. Descriptions of Terms Specific to This Method
3.2.1 activated—refers to the state of a qualitative detector's
response when indicating the presence of hydrocarbons.
3.2.2 detection time—elapsed time from a detector's first contact
with test product to an output that is within 95% of full scale or activated.
3.2.3 fa77 t/me—elapsed time after the detector is removed from test
hydrocarbon liquid until its output returns to within 5% of its original
baseline level or there is no detectable signal output.
3.2.4 non-activated—refers to the state of a qualitative detector's
response when indicating that no hydrocarbons are detected.
3.2.5 probe—component of a detection system that must come into
contact with petroleum product before the product can be detected.
3.2.6 qualitative responses—type of detector response that indicates
only the presence or absence of hydrocarbons without determining the specific
hydrocarbon thickness.
3.2.7 quantitative responses—type of detector response that
quantitates the concentration of the hydrocarbon present.
2
-------
3.2.8 relative accuracy—absolute mean difference between a group of
measured values and the true value, plus the 2.5% error confidence
coefficient, divided by the true value. Relative accuracy is a measure of the
maximum expected bias (without regard to sign) for a series of measurements.
3.2.9 response—detector's indication of the presence of petroleum
hydrocarbons. Responses can be qualitative or quantitative.
3.2.10 response time—general term that refers to the more specific
terms of detection time and fall time.
3.2.11 test product—commercial or synthetic gasoline used to
characterize detector performance.
4. Summary of Test Method
4.1 Detector probes are supported in a container that has a layer of
liquid hydrocarbon test product on water.- Detectors are tested five times at
each test product thickness. Test product thicknesses are 0.040 cm, 0.32 cm,
and 0,.64 cm. Detector response is monitored for up to 24 hours.. The
performance of detectors is tested with commercial unleaded gasoline and
synthetic gasoline made according to Method X 0004 with compounds that are
representative of components of gasoline.
5. Significance and Use
.' .5.1 For liquid-phase petroleum hydrocarbon detectors, accuracy is a
measure of how well the detector's output compares to a known thickness of
hydrocarbon product on water. Accuracy measurements provide a means for
>-- . ...
estimating the reliability of a detector. .
5.2 Precision is the degree of agreement of repeated measurements of
the same parameter. Precision estimates reflect random error and are not
affected by bias. In this method, precision is expressed in terms of the
percent coefficient of variation. ,
-,•'•''. 3 • ' .. ' •- '
-------
5.3 In addition to these primary performance characteristics,
procedures for estimating the magnitude and direction of measurement bias', are
presented. Bias is the systematic error inherent in a method, which may be
positive or negative. In this method, bias is expressed as the signed percent
difference between the average measured value for a series of tests and the
true value. ,
5.4 Response time is the general term that refers to a combination •
of the more specific terms detection time and fall time. Detection time is
the elapsed time from a detector's first contact with a given thickness of
petroleum product until it reaches 95% of its full-scale signal output or to
an activated response. Fall time is the elapsed time after the detector is
removed from contact with petroleum hydrocarbon until the detector output
returns to a stable baseline response.
5.5 Results obtained using this method will permit the most
advantageous use of a detector. Weaknesses as well as strengths of the
instrument should become apparent. It is not the interest of this method to
compare similar detectors from different manufacturers, but to enable the user
to choose a suitable detector.
6. Interferences
6.1 Conditions that can cause interferences with this method include
temperature changes, high temperatures, test product evaporation/vibrations,
and test products containing water-miscible substances (e.g., alcohol). To
avoid these conditions, tests should be conducted at constant (±3°C), normal
laboratory temperatures on a stable base, Evaporation can be reduced by
covering the test apparatus and maintaining a seal around the detector probe
and test container openings.
-------
6.2 It is difficult to verify the thickness of thin layers of liquid
hydrocarbon product on water. For this* reason, product thicknesses are
determined by calculation. For these calculations to be accurate, variations
' * ' , • ' j
in the dimensions of the test container should be minimal. Use containers
that have straight vertical sides with no visible protuberances or
indentations.
7. Apparatus
7.1 Test Container—All tests should be performed in non-reactive
containers with constant proportions. The containers hold water and test
product during testing. An example test container is depicted in Figure 1.
7.1:1 The test container should be cylindrical so that the thickness
of test product added to the container can be accurately calculated. The
container walls should be perpendicular to its bottom with no visible
protuberances or indentations throughout their height. The container diameter
should not deviate by more than 0.1 mm (0.004 in.) throughout the region where
hydrocarbon liquid will be contained during testing. The bottom of the
container can be rounded as long as the rounded portion does not extend above
the water level during tests. Also, the container walls should not deflect
.' ' " . - ' , i
visibly when filled with liquid. A typical container can be constructed by
welding a 6-inch outside diameter, Type 3041, seamless, stainless steel pipe/
section to a square plate of the same material, as depicted in Figure 1-. The
pipe and base plate sections should be at least 1/4-inch thick to minimize
/ • - ' - - '
warping during welding.
7.1.2 Container dimensions are dependent on detector probe dimensions
and operation. Containers with an inside diameter of 5-1/2 inches and a
height between 12 to 18 inches should be adequate for most detectors.
-------
7.1.3 The container and its base should be machined to accommodate a
thermocouple, a grounding wire, and a cover-grounding cable. Thermocouple
fittings can be accommodated in a 3/8-inch, NPT-threaded hole. The hole
should be located so that the thermocouple does not interfere with operation
of the detector probe. A grounding wire,and cover-grounding cable can be
accommodated by fitting the container base with two #10 screw holes.
7.2 Container Cover—A cover that fits snugly over the container and
supports the detector probe should be used. The cover should prevent vapor
loss from the container.
7.2.1 An appropriate cover can be made by machining a 1/4-inch thick
by 8-inch square, stainless steel plate. A 1/8-inch-deep groove should be cut
into one side of the plate. The width and diameter of the groove should allow
the plate to fit snugly onto the test container. The plate should have a hole
to accommodate the detector probe. The diameter of the probe .hole is
dependent on the dimensions of the probe. ,
7.2.2 Connect the cover to the container with a braided steel cable.
The cable grounds the cover to the container. Remove any plastic or rubber
insulation at the ends of the braided steel cable before fastening. A hole
for a #10 screw should be drilled and tapped into .the cover to fasten the
cable.
7.3 Timer—A timer that is accurate and precise to at least one
second per 10 minutes is required. Alternatively, a chart recorder or-other-
data acquisition'system can be used. If a chart recorder or other data
acquisition system is to be used, the timer is not required. If,used, the
recorder or data acquisition system timing must be accurate and precise to at
least one second per 10 minutes.
-------
7.4 Electronic Recorder—k chart recorder or other electronic data
acquisition system may be used if it is compatible with the specific detector
being evaluated. The output of the data recorder should be accurate and pre-
cise to ±2% over the range of output from a quantitative-output detector. A
data recorder used with a qualitative-output detector must unambiguously
identify activated and inactivated states.
7.5 Thermocouple—A thermocouple and temperature readout, or
equivalent, that responds from 0°C to 40°C and is accurate and precise to
within 18C over this range is needed.
7.6 Thermometer—-ASTM Solvents Distillation Thermometer having a
range from -2°C to 52°C and conforming, to the requirements for-Thermometer 37C
as prescribed in Specification E 1.
, 7.7 Str/ng-r-Approximately 30 cm of cotton or nylon string with a
diameter no greater than 0.2 cm may be needed to measure the cross-sectional
area of the detector. Alternatively, calipers with 0.1-mm or finer gradations
may be used.
7.8 Calipers—0.1 mm or finer gradations capable of measuring inside
diameter of test container.
8. Reagents and Materials
8.1 Purity of Reagents—Reagent grade chemicals shall be used in all
tests, except that commercial gasoline may be purchased from an ordinary
retail outlet. Unless otherwise indicated, it is intended that all reagents
conform to the specifications of the Committee on Analytical Reagents for the
American Chemical Society where such specifications are available.1 Other
"Reagent Chemicals, American Chemical Society Specifications," Am.
Chemical Soc., Washington, D.C. For suggestions; on the testing of reagents
not listed by the American Chemicals Society, see "Reagent Chemicals and
Standards," by Joseph Rosin, D. Van Nostrand Co., In., New York, NY, and the
"United states Pharmacopeia " .
-------
grades may be used, provided it is first ascertained that the reagent is of
sufficiently high purity to permit its use without lessening the accuracy of
determination.
8.2 Purity of Water—Unless otherwise indicated, references to water
shall be understood to mean drinking water or other relatively pure water with
an electrical conductivity (Method D;1125) of at least 50 ^mhos/cm.
8.3 Commercial Gasoline—Commercial gasoline test product shall be
unleaded regular or premium gasoline that is purchased at a retail outlet.
The gasoline shall contain less than, 2% water-miscible substances.
(Danger—Gasoline is extremely flammable. Vapors are harmful if inhaled. See
Annex Al.l. Leaded gasoline should not be used because there are significant
additional hazards associated with its handling and disposal).
8.4 Synthetic Gaso1ine—Synthetic gasoline, as used in this method,
is a 12-component mixture that is roughly representative of automotive
gasoline prepared according to Method X 0004. fDanger—Synthetic gasoline is
extremely flammable. Vapors are harmful if inhaled. See Annex Al.2).
8.5 Other Test Products—This test method can also be used with
other non-viscous, water-immiscible liquids. The method, however, does not
directly address use of liquids other than commercial gasoline, arid the
synthetic gasoline described herein. The suitability of this method with
regard to other substances should be ascertained before this method is used
with those liquids. ,
9. Calibration and Standardization
9.1 Chart Recorder or Other Data Recording System—If used, a chart
recorder or other data recording system should be calibrated along with the
detector. The data recording system should be calibrated according to
8
-------
instructions from its manufacturer and the detector manufacturer. Also, any
recording device .should be compatible with the detector being investigated.
Consult specifications from the manufacturers of the recording device and the
detector.
9.2 Detector—Because of wide design variability among different
petroleum detectors, it is impossible to give complete calibration
instructions for all possible detector designs. Calibrate all detectors
according to manufacturer instructions.
9.3 Probe Cross-Sectional Area—Estimate the area of the cross-
section of the probe that will be parallel to and at the same level as test
product.
9.3.1 The area of cylindrical or similarly shaped, many-sided
polygonal probe cross-sections should be measured by wrapping a thin string
around the detector cross-section that is parallel to and at the level of test:
product. Fit one loop of string tightly around the detector, and cut or mark
*• ' '
the string where it begins to overlap itself. The length of one Loop of •
string around the detector is the perimeter. Derive the detector radius from
the following equation: <
Radius, cm = p/(2 x 3.1416) '• . 0)
where:
p = length of perimeter, cm. . ,
The following formula should be used to calculate the area of a round cross-
section:
Cross-sectional area> cm = 3.1416 x r ~(2)
where:
r = radius of the cross-section, cm.
' • -'. • '. '9 . ' '
-------
9.3.2 Alternatively, the probe diameter can be measured with calipers.
The radius, which is half the diameter, can then.be used to calculate the
cross-sectional area by Equation 2.
9.3.3 Determine the area of rectangular cross-sections, by measuring
the length and width of the detector cross-section that is parallel to and at
the level of test product. Calculate the cross-sectional area using the
following equation:
Cross-sectional area, cm2 = 1 x w (3)
where:
1 « length, cm; and - , .
w » width, cm.
9.3.4 For detectors that have more than one part that will displace
test product, calculate the cross-sectional area for each part. The total
area for the probe is the sum of cross-sectional areas for individual parts.
9.3,5 For irregularly shaped probes, determine the cross-sectional
area by immersing the probe in water. Make two marks part-way up one side of
a transparent container that will contain the detector probe. The marks
should be 1.27 cm apart, and the container should have vertical walls 'in1 the
region where the marks are made. Add water to the lower mark. Using a buret,
determine the volume of water required to reach the upper mark to the nearest
0.1 ml. Empty the container, and suspend the detector probe inside the con-
tainer. The region of the probe that will contact test product should be
centered between the container marks. Again, fill the container to the lower
mark with water and determine the volume of water to reach the upper mark.
Calculate the probe cross-sectional area using the following equation:
Probe area, cm2 = (V, - Vp)/1.27
10
(4)
-------
where:
Vi = Volume between marks without probe, ml;
Vp = Volume between marks with probe, mL; and
i
1.27 = height of column of water displaced; cm.
9.4 Thermocoup1e—Perform side-by-side multipoint calibrations for
each thermocouple used in the test procedure in a 1-L glass beaker filled with
water. The reference thermometer should be an ASTM Solvents Distillation
Thermometer having a range from -2eC to 52'C and conforming to the
requirements for Thermometer 37C as prescribed in Specification E 1. The
levels tested are low (room temperature - 10°C), room temperature, and high _
(room temperature; + 108C).
9.4.1 Insert both the thermocouple and reference thermometer into the
beaker of water and add small quantities of ice. Allow the ice to melt and
the temperature to stabilize. Continue adding ,ice until a steady-state
reading (±0.5eC over two minutes) of room temperature - 10*C (±2°C) occurs.
9.4.2 Repeat this procedure using room temperature water (15*C to
30*C), and room temperature. + 10*C (±2*C) water. If the temperature
difference is more than 1*C, either repeat the test, with the same thermocouple
or replace the thermocouple and repeat the test until it is acceptable.
• • "J , r
9.4.3 Perform thermocouple calibration at the onset of testing and at
least once a year.
9.5 Test Container Xlrea—Accurately measure the inside diameter of
>— ' , . '
the test container to 0.1 mm or finer. Calculate the container cross-
i
sectional area with the following formula:
Area, cm2 = 0.785 x dr2 (5)
c
where:
dc = test container inside diameter, cm.
11
-------
10. Procedure
10.1 Tes.t Series—The detector should be tested a minimum of five
times for each combination of test product and hydrocarbon layer thickness
listed in Table 1.
Note I—Do not reuse product-soluble components that have been exposed
to product. Instead, replace these components.
10.1.1 Perform tests in a random order such that variables of test
product and hydrocarbon layer thickness are,isolated.
10.2 Add an appropriate volume of water to the container. The volume
of water added must allow the detector probe to be fully functional. If the
probe operational characteristics'do not place limitations on the water
volume, then add 2 L of water to the container. The water should be within
2'C of room temperature, which should be between 15°C and 28°C.
10.3 Place the cover on the; container and connect the steel braided
cable between the cover and the container.
10.4 Mount the probe in the test container so that it is fully
functional and forms a tight seal with the cover. The manufacturer's
specifications should give details on the placement of the detector in a well
with regard to the bottom of the well and in relation to the depth of water in
the well. Place the detector in the test container as if the test container
were a we!1.
10.5 Electrically connect the test container to an earth ground to
>•- .
dissipate static electricity. ,
10.6 Connect the detector output to a chart recorder or other data >
acquisition system if one is being used. All connections should be in
compliance with specifications from the manufacturers of the detector and the
data recording system. ••'.,-'.
12
-------
10.7 Calibrate the detector if necessary. Many detectors, such as
product-soluble detectors, do not require any calibration. Perform
calibrations, if necessary, according to manufacturer recommendations.
Calibration may need to occur before mounting in the test container. If a
data recording system is being used, it should be calibrated with the
detector. Calibrate the data acquisition system according to manufacturer
instructions. , ,
10.8 Blank Test—Perform a blank test by monitoring the detector
output for 30 minutes while the probe is in water.
'" ' , *
10.8.1 For quantitative detectors, record the stable output level from
the detector at the end of the 30-minute period. If the detector output is
not stable after 30 minutes, wait until the output becomes stable. Record the
stable output level.
10.8.2 If a qualitative detector goes to an activated state during the
blank test, correct the detector malfunction or replace it with a properly
operating detector.-
10.9 Determine the amount of product to add to the water. It is
important to accurately calculate the amount of product that is needed to form
a product layer. Calculate the volume of product to add to the test container
with the following equation: /• .
Volume, ml = t x (ac - ad) ' .• ' . (6)
where:
1 >.. . ' - , . , ' .'.,''
t = desired product thickness in cm;
ac - test container cross-sectional area in cm2; and
a,i = estimated detector cross-sectional area in cm2.
13
-------
10.10 Remove the container coyer with the attached detector probe from
the container, and add the appropriate volume of test product with a glass
buret. Pour the test product without splashing or contacting the container
walls.
10.11 Cover the top of the test container immediately to reduce test
product evaporation and to reduce ambient vibrations from air currents.
10.12 Data Recording—Begin monitoring the detector response
immediately after replacing the probe and cover. Do not stir or otherwise
disturb the contents of the test container. The detector output may be
recorded by hand, with a chart recorder, or with another type of electronic
data acquisition system. The nature of response monitoring is dependent on
whether the device signal is quantitative or qualitative.
10.12.1 For quantitative detectors, monitor the output signal at least
until the signal becomes stable or 24 hours elapses, whichever 1s.shorter.
Record the detector output at the end of the test period.
10.12.2 For qualitative detectors, monitor the detector output at
least until it activates or 24 hours elapse, whichever is shorter. Record the
detector output state (activated or not activated) at the end of the test
period.
10.13 Detection Time—If the detector gives a positive response within
24 hours, the elapsed time between when the probe was pl.aced in the container
and when the detector responded is the detection time. The nature of a
response is dependent on whether a detector gives quantitative or qualitative
output.
10.13.1 The period for detection, time of quantitative detectors is
from introduction of the detector probe into the test product to the time the
14
-------
' 'detector reaches 95% of its final stable output. Calculate the 95% of final
stable output level from the,following equation:;
-. f
High level output, >cm = BL + (Hi - BL) x 0.95 (7)
where:
BL = stable baseline output, cm; and
HL = stable high level output,cm.
10.13.2 A positive response for qualitative detectors occurs when the
detector output goes from an inactivated state to an activated state.
10.13.3 If the detector gives a response within 24 hours, report the
elapsed time between when the detector probe was added to the container with
, test product and when the detector responded as the detection time.
10.13.4 , Detectors with lower detection limits above a particular test
product thickness do not need to be tested for the full 24-hour period. They
should be tested for at least five times their maximum expected detection
time.
10.14 Record water and product temperature in the test container in
degrees Centigrade. If the change in temperature since the first temperature
measurement is more than 8"C, repeat the test.
10.15 Fall Time—Test fall time by raising the detector probe from the
test container, rinsing the probe with fresh water, draining and refilling the
test container, and replacing the detector in the container. Alternatively,
the probe can be immersed in a second clean container. The test container
cover may need to be removed to accommodate removal of the detector probe.
15
-------
10.15.1 Fall time is not applicable to some detectors such as product-
soluble detectors, and qualitative-output detectors that did not activate in
response to addition of product to the test container.
10.15.2 Start the timer or mark the beginning of the fall time test on
the recording system when the detector is lifted from the liquid.
10.15.3 Support the detector probe above any counter tops while the
test container is being drained.
10.15.4 After the initial liquid is completely drained from the
container, rinse the container with fresh water until there is no evidence of
product residue. An acetone rinse will facilitate removal of test product.
Draining and rinsing should be completed within two minutes. . Alternatively,
fall time can be tested by rinsing the probe with water and then immersing the
probe in water in a second container.
10.15.5 When there is no evidence of test product in the container,
add fresh water to the container and replace the detector and container cover.
10.15.6 Monitor the detector output for fall time response. The
nature of a fall time response is dependent on whether a detector gives
quantitative or qualitative output. Fall time response for a quantitative
detector is when the detector output returns to within 5% of its original
stable baseline level. Calculate the 5% stable baseline level according to
the following equation:
5% Stable baseline output, cm = BL + (HL - BL) x 0.05 (8)
where: ,
BL = stable baseline output, cm; and
HL = stable high level output, cm.
16
-------
Fall time response for a qualitative detector is when the detector output goes
from an activated state to an inactivated state.
10.15.7 Continue fall time monitoring for up to 24 hours.
10.15.8 If the fall time response occurs before the detector has been
returned to the container, fall time is less than the amount of elapsed time
between when the timer was activated and the fall time response occurred.
10.15.9 Some detectors may not return to baseline conditions when
submersed in water. Alternatively, test these detectors with their probes in
air.
11. Calculations .
11.1 Relative Percent Difference—Calculate relative percent
difference as follows:
Relative percent difference, % = 200 x [(Vj - V2)/'(Vj + V2)] (9)
where:
Vj = larger value, cm; and
V2 - smaller value, cm.
11.2 Coefficient of Van at ion—Calculate the coefficient of variation
as follows:
Coefficient of variation, %= (s/X) x 100 (10)
where:
s = standard deviation of n values (n-1 degrees of freedom),cm; and
X = mean of n values, cm.
.17 :.'••'.'•'
-------
11.3 Accuracy—Calculation for accuracy is dependent on the type of
output that a detector produces.
11.3.1 Quantitative detectors—Accuracy for quantitative detectors is a
function of systematic error (bias) and random error (precision). Calculate
relative accuracy (RA) of a set of data as follows:
Relative accuracy, % = (|d| + |cc|) / Vr x 100 (11)
where:
Vr - reference (theoretical) value, cm; -
d = arithmetic mean of the difference of a data set, Equation 12; and
cc - 2.5% error confidence coefficient (one tailed, Equation 13), cm.
11.3.2 Mean difference—Calculate the arithmetic mean of the difference
(d) of a data set as follows:
n
Mean difference, cm = 1/n 2 d, . (12)
i-i
where:
di = measured response - theoretical response, cm.
11.3.3 Confidence coefficient—Calculate the one-tailed 2.5% confidence
coefficient (cc) as follows:
>•- . , • '
Confidence coefficient, cm = t0 975 x s/7n (13)
where:
s » the standard deviation (n-1) of the data set, cm; and
tog75 - 2.5% t value from Table 2; and
18 _
-------
n = number of tests for a test product at a particular thickness.
11.3.4 Qualitative detectors—Use the following formula to calculate
the accuracy of qualitative detectors:
Accuracy, % = 100 x (r^/n) (14)
where: ,
rp = number of positive responses; and
n = number of tests for a particular test product at a particular thickness.
11.4 fl/as—Bias for quantitative detectors is calculated as follows:
Bias, % = 100 x [(V0 - Vr)/Vr] (15)
where:
Vj = the individual response to test product, cm;
Vr = the reference (theoretical) value, cm; and
n
V0 = the average observed value, 1/n 2 V,, cm;
i-l
where:
n = the number of tests with a particular test product at a particular test
product thickness. ,
11.5 Detection I/me—-Calculate detection time according to the
following formula:
Detection time.= T2 - Tj (16)
where: ,
Tj = clock time when liquid was first added to test container; and
T2 = clock time when detector output went from an inactivated state to
19
-------
an activated state for a qualitative detector or from a baseline reading to 5%
of stable high level output for a quantitative detector.
11.6 Fall Time—Calculate fall time according to the following
formula:
Fall time = T -
(17)
where:
Tj » clock time when detector was removed from test container; and
T2 = clock time when detector output went from an activated state to an
inactivated state for a qualitative detector or from a high level reading to
within 95% of stable baseline level output for a quantitative detector.
12. Report
12.1 Use the form displayed in Figure 2 to report results. Report
the following information: .
12.1.1 Detector type—Report whether the detector was a quantitative or
qualitative type.
12.1.2 Accuracy—For quantitative detectors, report relative accuracy
for both test products at every test product thickness (0.040, 0.32, and 0.64
cm) according to Equation 11. Report accuracy for qualitative detectors
according to Equation 14.
12.1.3 Precision—Precision for quantitative detectors is defined as
the percent coefficient of variation. Use Equation 10 to calculate
coefficient of variation. For quantitative detectors, report precision as the
percent coefficient of variation for both test products at every test product
thickness (0.040, 0.32, and 0.64 cm). Precision is not applicable to
20
-------
qualitative detectors, and shall be reported as "NA", not applicable, for
these detectors.,. ,
12.1.4 Bias—Bias at a particular thickness for a particular test
product is the percent difference between the average detector output for a
series of tests and the actual thickness of test product, Equation 15. For
quantitative detectors, report bias for both test products at every test
product thickness (0.040, 0.32, and 0.64 cm). Bias is not applicable to
qualitative detectors, and shall be reported as "NA," not applicable, for
these detectors.
12.1.5 Detection Time—Report detection time in the largest convenient
units (seconds, minutes, or hours) for both test products at every test
product thickness (0.040, 0.32, and 0.64 cm). If detector response is
immediate, report detection time as "<1 second." If the detector does not
respond within 24 hours, report detection time as "No response in 24 hours."
12.1.6 Fall Time—Report fall time in the largest convenient units
(seconds, minutes, or^hours) for both test products at every test product
thickness (0.040, 0.32, and 0.64 cm). Fall time is not applicable to some
detectors such as product-soluble detectors and qualitative detectors that did
not activate in response to addition of product to the test container. For
these detectors, report fall time as "NA," not applicable. Also, record the .
reason(s) why fal.l time determination is not applicable. If the detector does
not respond within 24 hours, report fall time as "No response in 24 hours."
13. Precision and Bias
13.1 Precision—Jhe precision of the procedure in Test Method X 0000
for measuring accuracy and response time for liquid-phase out-of-tank
petroleum detectors is being determined.
21
-------
13.2 B/as—Since there is no accepted reference material suitable for
determining the bias for the procedure in Test Method X 0000 for measuring
accuracy and response time for liquid-phase out-of-tank .petroleum detectors,
no statement on bias is being made.
22
-------
TABLE 1. Test Product and Thickness Combinations
Test Product , Layer Thickness, cm
commerci al gasqli ne 0.040
commercial gasoline 0.32
commercial gasoline 0.64
synthetic gasoline 0.040
synthetic gasoline >. . 0.32
synthetic gasoline . 0.64
23
-------
TABLE 2. 2.5% T Values^1
o
• 1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
40
60
120
— : . • ™ '
12.706
4.303
3.182
2.776
2.447
2.571
2.365
2.306
2.262
2.228
2.201
2.179
2.160
2.145
2.131
2.120
2.110
2.101
2.093
2.086 -
2.080
2.074
2.069
2.064
2.060
2.056
2.052
2.048
2.045
2,042
2.021
2.000
1.980
1.960
ATaken from CRC Standard Mathematical Tables, 26th ed. CRC Press, Inc. Boca
Raton, FL, 1981.
24
-------
1/4 " GROOVE
COVER
PROBE
OPENING
THERMOCOUPLE
PROBE
12 IN. OR 18 IN.
GROUNDING |
LEAD
, FIG1. Test Container and Cover
25
-------
Detector name:.
Detector type:
Test Product
Quantitative
Qualitative
Thickness, on Accuracy, % Precision, % Bias. % Detection Time Fall Time
ro
01
commercial gasoline 0.040
commercial gasoline 0.32
commercial gasoline 0.64
synthetic gasoline 0.040
synthetic gasoline 0.32
synthetic gasoline 0.64
COMMENTS:
FIG 2. Data Recording Form
-------
ANNEX
.-'•., (Mandatory Information)
Al.l Gasoline (including Leaded Gasoline)
Danger—Extremely flammable. Vapors harmful if inhaled.
Vapors may cause flash fire.
Contains toxic lead antiknock components. Harmful if absorbed through
skin.
Keep away from heat, sparks, and open flames.
Keep container closed. .
Use with adequate ventilation. Avoid buildup of vapors and eliminate
all sources of ignition especially nonexplosion-proof electrical apparatus and
heaters. ^
Avoid prolonged breathing of vapor or spray mist. Avoid prolonged or
repeated skin contact. .
A1.2- Synthetic Gasoline Mixture
Danger—Extremely flammable. Vapors harmful if inhaled. Vapors may
cause flash fire. Contains toxic benzene and other hydrocarbon substances;
(See Method X 0000).
Harmful if absorbed through skin.
Keep away from heat, sparks, and open flames.
Keep container closed. Use with adequate ventilation.
Avoid buildup of vapors and eliminate all sources of ignition,
especially nonexplosion-proof electric apparatus and heaters.
Avoid prolonged breathing of vapors or spray mist.
27
-------
Avoid prolonged or repeated skin contact.
28
-------
Designation: X 0002
Standard Test Method for
SPECIFICITY FOR LIQUID-PHASE OUT-OF-TANK PETROLEUM DETECTORS
1. Scope
1.1 This test method covers determination of specificity of liquid-
phase out-of-tank petroleum hydrocarbon leak detectors that utilize ground
water monitoring wells.
1.2 This method is applicable to only the components associated with
detection of Iiquid1phase petroleum releases for detection systems utilizing
multiple operating principles. ,
1.3 This standard may involve hazardous materials, operations, 'arid
equipment. This standard does not purport to address all of the safety
problems associated with its use. It is the responsibility of the user of
this standard to establish appropriate safety and health practices and
determine the applicability of regulatory limitations prior to use.
2. Referenced Documents
2.1 ASTM Standards:
D 1125 Standard Test Methods for Electrical Conductivity and
Resistivity of Water
1 ••-'-.'.
-------
E 1 Standard Specification for ASTM Thermometers
E,.456 Standard Terminology Relating to Statistics
2.2 EPA Leak Detector Standards:
X 0004 Standard Practice for Preparation of Synthetic Gasoline
for Testing Out-of-Tank Petroleum Detectors.
3. Terminology
3.1 Definitions — For formal definitions of statistical terms, see
Terminology E 456.
3.2. Descriptions of Terms Specific to This Method
3.2.1 activated—refers to the state of a qualitative detector's
response when indicating the presence of hydrocarbons.
3.2.2 non-activated—refers to the state of a qualitative detector's
response when indicating that no hydrocarbons are detected.
3.2.3 probe—component of a detection system that must come into
contact with petroleum product before the product can be detected.
3.2.4 qualitative responses—type of detector response that indicates
only the presence or absence of hydrocarbons without determining the specific
hydrocarbon thickness.
3.2.5 quantitative responses—type of detector response that
quantitates the concentration of the hydrocarbon present.
3.2.6 response—detector's indication of the presence of petroleum
hydrocarbons. Responses can be qualitative or quantitative.
3.2.7 specificity—ability of a detector to respond to various
* . .
substances. ,
3.2.8 test product—commercial or synthetic gasoline used to
characterize detector performance.
-------
4. Summary of Test Method
4.1 Detector probes are supported in a container that has a 1.27-
cm-thick layer of liquid hydrocarbon test product on water. .Probes are
exposed to each of seven different test products. Detector response is
monitored for up to 24 hours.
5. Significance and Use
5.1 For liquid-phase petroleum hydrocarbon detectors, specificity is
a measure of how sensitive a detector is to different test products.
Specificity measurements provide a means for estimating the suitability of a
detector for different stored products. '
5.2 Results obtained using this method will permit the most
advantageous use of a detector. Weaknesses as well as strengths of the
instrument should become apparent. It is not the interest of this method to
compare similar detectors of different manufacture, but to enable the User to
choose a suitable detector.
6. Interferences
6.1 Conditions that can cause interferences with this method include
temperature changes, high temperatures, test product evaporation, vibrations,
and test products containing water-miscible substances (e.g., alcohol). To
avoid these conditions, tests should be conducted at constant (±3eC), normal
laboratory temperatures on a stable base. Evaporation can be reduced by
covering the test apparatus and maintaining a seal around the detector probe
"•-•'. -' . ' • . . - • '
and test container openings. , ,
6.2 It is difficult to verify the thickness of thin layers of liquid
hydrocarbon product on water. For this reason, product thicknesses are .
determined by calculation. For these calculations to be accurate, variations
in the dimensions of the test container should be minimal. Use containers
. . '3 •- - • . : •
-------
that have straight vertical sides with no visible protuberances or
indentations. ,.
7. Apparatus
7.1 Test Container—All tests should be performed in non-reactive
containers with constant proportions. The containers hold water and test
product during testing. An example test container is depicted in Figure 1.
7.1.1 The test container should be cylindrical so that the thickness
of test product added to the container can be accurately calculated. The
container walls should be perpendicular to its bottom with no visible
protuberances or indentations throughout their height. The container diameter
should not deviate by more than 0.1 mm (0.004 in.) throughout the region where
hydrocarbon liquid will be contained during testing. The bottom of the
container can be rounded as long as the rounded portion does not extend above
the water level during tests. Also, the container walls should not deflect
visibly when filled with liquid. A typical container can be constructed by
welding a 6-inch outside diameter, Type 304L, seamless, stainless steel pipe
section to a square plate of the same material, as depicted in Figure i: The
pipe and base plate sections should be at least 1/4 inch thick to minimize
warping during welding.
7.1.2 Container dimensions are dependent on detector probe dimensions
and operation. .Containers with an inside diameter of 5-1/2 inches and a
x
height between 12 to 18 inches should be adequate for most detectors.
7.1.3 "The container and its base should be machined to accommodate a
thermocouple, a grounding wire, and a cover-grounding cable. Thermocouple
fittings can be accommodated in a 3/8-inch, NPT-threaded hole. The hole
should be located so that the thermocouple does not interfere with operation
-------
of the detector probe. A grounding wire and cover-grounding cable can be
accommodated by..fitting the container base with two #10 screw holes.
7.2 Container Cover—A cover that fits snugly over the container and
supports the detector prpbe should be used. The cover should prevent vapor
loss from the container.,
7.2.1 An appropriate cover can be made by machining a 1/4-inch thick
by 8-inch square, stainless steel plate. A 1/8-inch-deep groove should be cut
into one side of the plate. The width and diameter of the groove should allow
the plate to fit snugly onto the test container. The plate should have a hole
to accommodate the detector probe. The diameter of the probe hole is
dependent on the dimensions of the probe.
7.2.2 Connect the cover to the container with a braided steel cable.
The cable grounds the cover to .the container. Remove any plastic or rubber
insulation at the ends of the braided steel cable before fastening. A hole
for a #10 screw should be drilled and tapped into the cover to fasten the
cable.
7.3 Timer—A timer that is accurate and precise to at least one
second per 10 minutes is required. Alternatively, a chart recorder or other
data acquisition system can be used. If a chart recorder or other data
acquisition system is to be used, the timer is not required. If used, the
recorder or data acquisition system timing must be accurate and precise to at
least one second per 10 minutes.
7.4 Electronic Recorder—A chart recorder or other electronic data
' - " _ '"*-••; ^
acquisition system may be used if it is compatible with the specific detector
being evaluated. The'output of the data recorder should be accurate and pre-
cise to ±2% over the range-of output from a quantitative-output detector. A
-------
data recorder used with a qualitative-output detector must unambiguously
identify activated and inactivated states. . ,
7.5 Thermocouple—A thermocouple and temperature readout, or
equivalent, that responds from O'C to 40°C and is accurate and precise to
within 16C over this range is needed.
7.6 Thermometer—ASTM Solvents Distillation Thermometer having a
range from -2eC to 52°C and conforming to the requirements for Thermometer. 37C
as prescribed in Specification El. ', .
7.7 String—Approximate!y 30 cm of cotton or nylon string with a
diameter no greater than 0.2 cm may be needed to measure the cross-sectional
area of the detector. Alternatively, calipers with 0.1-mm or finer .gradations
may be used.
7.8 Calipers—O.I mm or finer gradations capable of measuring.inside
diameter of test container. , .
8. Reagents and Materials
8.1 Purity of fleagents—Reagent grade chemicals shall be used in all
tests, except that commercial fuels may be purchased from ordinary retail
outlets. Unless otherwise, indicated, it is intended that all. reagents conform
to the specifications of the Committee on Analytical Reagents for the American
Chemical Society where such specifications are available.1 Other grades may
be used, provided it is first ascertained that the reagent is of sufficiently
high purity to permit its use without lessening the accuracy of determination.
^'Reagent Chemicals, American Chemical Society Specifications," Am.
Chemical Soc., Washington, D.C. For suggestions on the testing of reagents
not listed by the American Chemicals Society, see "Reagent Chemicals and
Standards," by Joseph Rosin, D. Van Nostrand Co., In., New York, NY, and the
"United States Pharmacopeia." .
-------
8.2 Purity of Water—Unless otherwise indicated, references to water
shall be understood to mean drinking water or other relatively pure water with
an electrical conductivity (Method D 1125) of at least 50 /anhos/cm.
8.3 Commercial Gaso77/7e—Commercial gasoline test product shall be
unleaded regular or premium gasoline that is purchased at a retail outlet.
The gasoline shall contain less than 2% water-miscible substances.
(Danger—Commercial gasoline is extremely, flammable. Vapors are''harmful if
inhaled. Leaded gasoline should not be used because there are significant
additional hazards associated with its handling and disposal. See Annex A1..1.
8.4 0/ese7 Fuel— Grade 2 automotive dieseV' fuel test product shall
be purchased at a retail outlet. The fuel shall contain less than 2% water-
miscible substances. (Danger—Diese1 fuel is flammable. Vapors are harmful
if inhaled. See Annex A1.2.}
8.5 n-Hexane (C6H12). (Danger--n-Hexane is extremely flammable.
Vapors are harmful if inhaled. See Annex A1.3.)
8.6 Jet Fuel—Jet-A jet fuel test product shall be purchased at a
retail outlet. The fuel shall contain less than 2% water-miscible substances.
(Danaer—Jet fuel is flammable. Vapors are harmful if inhaled. See Annex
A1.4.) .
8.7 Synthetic Gasoline—Synthetic gasoline, as used in this method,
is a 12-component mixture that is roughly representative of automotive
gasoline prepared according to Practice X 0004. (Danger—Synthetic gasoline
is extremely flammable. Vapors are harmful if inhaled. See Annex A1.5.)
.8,8 Toluene (CH3C6H5). (Danger—To 1 uene is extremely flammable.
Vapors are harmful if inhaled. See Annex A1.6.)
8.9 Xylene(s) [2(CH3)C6H4]. (Danger—Xv 1 ene is flammable. Vapors
are harmful if inhaled. See Annex A1.7.)
: '- ' ' 7 '".''•"
-------
9. Calibration and Standardization
9.1 Chart Recorder or Other Data Recording System—If used, a chart
recorder or other data recording sy;stem should be calibrated along with the
detector. The data recording system should be calibrated according to
instructions from its manufacturer and the detector manufacturer. Also, any
recording device should be compatible with the detector being investigated.
Consult specifications from-the manufacturers of the recording device and the
detector.
9.2 Detector—Because of wide design variability among different
petroleum detectors, it, is impossible to give complete calibration
instructions for all possible detector designs. Calibrate all detectors
according to manufacturer instructions.
9.3 Probe Cross-Sectional Xlrea—Estimate the area of the cross-
section of the probe that will be parallel to and at the same level as test
product.
9.3.1 The area of cylindrical or similarly shaped, many-sided
polygonal probe cross-sections should be measured by wrapping a thin string
around the detector cross-section that is parallel to and at the level of test
product. Fit one loop, of string tightly around the detector, and cut or mark
the string where it begins to overlap itself. The length of one loop of
string around the detector is the perimeter. Derive the detector radius from
the following equation:
Radius, cm = p/(2 x 3.1416) , (1)
where:
p « length of perimeter, cm.
The following formula should be used to calculate the area of a round cross-
section: .
8
-------
Cross-sectional area, cm2 = 3.1416 x r2 (2)
where:
r = radius of the cross-section, cm.
9.3.2 Alternatively, the probe diameter can be measured with calipers.
The radius, which is half the diameter, can then be used to calculate the
cross-sectional area by Equation 2.
9.3.3 Determine the area of rectangular cross-sections by measuring'
the length and width of the detector cross-section that is parallel to and at
the level of test product. Calculate the cross-sectional area using the
following equation:
Cross-sectional area, cm2 = T x w (3)
where: ,
1 = length, cm; and
w = width, cm.
9.3.4 For detectors that have more than one part that will displace
' ' ' ' • * •
test product, calculate the cross-sectional area for each part. The total
area for the probe is the surn of cross-sectional areas for individual parts.
9.3.5 For irregularly shaped probes, determine the cross-sectional
area by immersing the probe in water. Make two marks part-way up one side of
a transparent container that will contain the detector probe. The marks
should be 1.27 cm apart, and the container should have vertical walls in the
region where the marks are made. Add water to the lower mark. Using a buret,
determine the volume of water required to reach the upper mark to the nearest
0.1 ml. Empty the container, and suspend the detector probe inside the con-
tainer. The region of the probe that will contact test product should be
centered between the container marks. Again, fill the container to the lower
' - . . 9 . • • .• -. . •''..-•
-------
mark with water and determine the volume of water to reach the upper mark.
Calculate the probe cross-sectional area using the following equation:
Probe area, cm2 = (V, - Vp)/1.27 (4)
where:
Vi s Volume between marks without probe, ml;
V = Volume between marks with probe, ml; and
1.27 - height of column of water displaced, cm.
9.4 Thermocoup7e—Perform side-by-side multipoint calibrations for
each thermocouple used in the test procedure in a 1-L glass beaker filled with
water. The reference thermometer should be an ASTM solvents distillation
thermometer having a range from,-2°C to 52°C and conforming to the .
requirements for Thermometer 37C as prescribed in Specification El. The
levels tested are low (room temperature - 10*C), room temperature, and high
(room temperature + 10'C).
9.4.1 Insert both the thermocouple and reference thermometer into the
beaker of water and add small quantities of ice. Allow the ice to melt and
the temperature to stabilize. Continue adding ice until a steady-state
reading (±0.5°C over two minutes) of room temperature - 10°C (±2°C) occurs.
9.4.2 Repeat this procedure using room temperature water (15°C to
30°C), and room temperature + 10°C (±2°C) water. If the temperature
difference is more than 1°C, either repeat the test with the same thermocouple
or replace the thermocouple and repeat the test until it is acceptable.
9.4.3 Perform thermocouple calibration at the onset of testing and at
least once a year.
10
-------
Results of U.S. EPA Standard Evaluation
Liquid-Phase Out-of-Tank Product Detectors
This form, documents the performance of the liquid-phase product detector described below. The
evaluation was conducted by the equipment manufacturer or a consultant to the manufacturer
according to the U.S. EPA's "Standard Test Procedure for Evaluating Leak Detection Methods:
Liquid-Phase Out-of-Tank Product Detectors."
Tank owners using this leak detection system should keep this form on file to prove compliance
with the federal regulations/Tank owners should check with state and local agencies to verify that
this form satisfies their requirements.
Method Description
Name • -' • •••••• ' _
Version.
Vendor
• • (street address)
' ~py)~~~~ (state) (zip) ' ~(phone)
Detector output type: D Quantitative LH Qualitative
r elector operating principle: D Electrical Conductivity D Thermal Conductivity
Snterface Probe D Product Permeable . D Product: Soluble • D Other
Detector sampling frequency: CD Intermittent EH Continuous
Evaluation Results
The detector above was tested for its ability to detect a layer of liquid floating on water, the
following parameters were determined:
Accuracy - How closely the product thickness, as measured by the detector, agrees with the
actual thickness.
Bias - Whether the method consistently over-estimates or under-estimates product thickness.
Not applicable to qualitative detectors.:
Precision - Agreement between multiple measurements of the same product thickness. Not
applicable to qualitative detectors.
Detection Time - Amount of time the detector must be exposed to product before it responds.
Fall Time - Amount of time that passes before the detector returns to its baseline reading after
the product is removed.
Lower Detection Limit - The smallest product thickness that the detector can reliably detect. To
meet federal performance standards, this must be less than 0.32 cm (1/8 inch).
Specificity - Indicates the accuracy of the detector in sensing several different liquids.
Liquid-Phase Product Detector - Results Form Page 1 of 2
-------
Liquid-Phase "Product Detector_
Version
Evaluation Results (continued)
> Compiled Tost Results (for tests conducted with 0.32 cm of floating product)
mmmercial Gasoline Synthetic Gasoline
Accuracy (%)
B!as*(%)
Precision* (%)
Detection Time (hh:mm:ss)
Fall Time (hh:mm:ss)
Lower Detection Limit (cm)
* Not applicable to qualitative detectors.
> Specificity Results (%)
Commercial gasoline
Synthetic gasoline
Diesel fuel •
Jet-A Jet fuel
n-Hexane
Toluene . _
Xylene(s)
> Safety disclaimer: This test procedure only addresses the Issue of the method's
ability to detect leaks. It does not testthe equipment for safety hazards.
Certification of Results
1 certify that the liquid-phase product detector was operated ace anting i to .the vendor
instructions and that the evaluation was performed according to the
Drocedure for liquid-phase out-of-tank product detectors except as noted on any attached _
sheets I also oeSy that the results presented above are those obtained during the evaluation.
(printed name)
(signature)
(organization performing eva.uai.6nF
(city, state, zip;
(phone numoeg
Liquid-Phase Product Detector - Results Form
age
-------
-------
------- |