 | |  | |  | |  | |  | |  |
1. The Time Frame for Conducting Stack Tests
2. Stack Test Waivers
3. Stack Test Notifications
4. Observation of Stack Tests
5. Representative Testing Conditions
6. Stoppages
7. Postponements
8. Test Reports
2. Stack Test Waivers
3. Stack Test Notifications
4. Observation of Stack Tests
5. Representative Testing Conditions
6. Stoppages
7. Postponements
8. Test Reports
2. STACK TEST WAIVERS
• Stack tests to determine and demonstrate initial compliance may be, in some instances, the
only test an emission unit will receive for an extended period of time. Therefore, all units should
be tested for initial compliance unless a waiver has been granted by the delegated agency
pursuant to 40 CFR §§ 60.8(b)(4), 61.13(h)(1)(iii), or 63.7(h). Waivers are granted only if the
owner or operator of a source has demonstrated by other means that the source is in compliance
with the applicable standard, or, under the MACT provisions, if the source is operating under an
extension of compliance pursuant to § 63.6(i), or has requested such an extension and the request
is under consideration by the delegated agency. The waiver regulations make clear that the
burden of proof is on the affected facility to justify the need for a waiver. Although the NSPS
and NESHAP programs do not specify what information is required as justification, the MACT
program in 40 CFR § 63.7(h)(3)(iii) states that the waiver application should include information
such as the "technical or economic infeasibility, or the impracticality, of the affected source
performing the required test."
• The primary issue of concern with respect to waiver requests is whether stack tests to
determine and demonstrate on-going compliance with emission limits should be waived under
the NSPS, NESHAP and MACT programs for units identical to a unit(s) that has been tested.
• Units, although identical in terms of design and control devices, may have process operations
that significantly alter their performance and ability to comply with the underlying regulatory
requirements on a continuing basis. Therefore, if the identical units have the ability to emit a
pollutant in excess of the prescribed emission limit, a stack test should not be waived without
adequate justification. However, a waiver may be appropriate on a case-by-case basis when
criteria such as the following are met:
(1) the units are located at the same facility;
(2) the units are produced by the same manufacturer, have the same model number or
other manufacturer’s designation in common, and have the same rated capacity and
operating specifications;
(3) the units are operated and maintained in a similar manner; and
(4) the delegated agency, based on documentation submitted by the facility,
(a) determines that the margin of compliance for the identical units tested is
significant and can be maintained on an on-going basis; or
(b) determines based on a review of sufficient emissions data that, though the
margin of compliance is not substantial, other factors allow for the determination
that the variability of emissions for identical tested units is low enough for
confidence that the untested unit will be in compliance.3 These factors may
include, but are not limited to, the following:
(i) historical records at the tested unit showing consistent/invariant
load;
(ii) fuel characteristics yielding low variability (e.g., oil)
and therefore assurance that emissions will be constant and
3 As a general matter, the greater the quantity of available emissions data, the smaller the
range of uncertainty about emissions and the more readily reviewing agencies can determine
precise levels of emissions variability. Under such circumstances, delegated agencies may have
greater assurance that compliance will be continuous even where the difference between actual
and permitted emission rates is relatively small.
Page 10
below allowable levels;
(iii) statistical analysis of a robust emissions data set
demonstrate sufficiently low variability to convey
assurance that the margin of compliance, though small, is
reliable.
• If a facility does not have the ability to emit a pollutant in excess of the prescribed emission
limit, waivers on a case-by-case basis may be issued for both initial and on-going compliance
stack tests. For example, a stack test waiver for identical units at a facility operating multiple
natural gas-fired boilers subject to a particulate matter standard generally would be appropriate.
• Waivers can be granted only by the appropriate delegated agency. See 40 CFR § 63.91(g).
See also, "How to Review and Issue Clean Air Act Applicability Determinations and
Alternative Monitoring," EPA 305-B-99-004, Section 4.2, pp.19-22 (February 1999). If the
delegated state/local agency has the authority to grant a waiver, it still should consult promptly
with EPA to promote national consistency.
3. STACK TEST NOTIFICATIONS
• The primary issue is what constitutes sufficient notification of a planned stack test under the
regulatory requirements. Sufficiency is defined to include both the timing of the notification, as
well as the content of the notification.
• Unless specified otherwise in the subpart, both the NSPS and NESHAP programs require at
least thirty (30) calendar days advance notice of a stack test [40 CFR § 60.8(d) and 40 CFR
§ 61.13(a) and ©], while the MACT program requires at least sixty (60) calendar days [40 CFR
§ 63.7(b)(1)]. The test date(s) and approximate start/end time of the test should be acceptable to
both the delegated agency and the facility to allow the delegated agency an opportunity to
observe the test, if desired. If for some reason the stack test must be delayed, facilities also are
required to provide notification of the delay. The time frame for such notifications differs under
each program. Under 40 CFR § 60.8(d), the facility is required to provide notification "as soon
as possible of any delay in the original test date, either by providing at least 7 days prior notice
of the rescheduled date of the performance test, or by arranging a rescheduled date with the
Administrator (or delegated state or local agency) by mutual agreement." Under 40 CFR
§ 63.7(b)(2), if the facility must delay the test due to "unforeseeable circumstances beyond [its]
control ", the facility must notify the "Administrator as soon as practicable and without delay
prior to the scheduled performance test date and specify the date when the performance test is
rescheduled." 40 CFR § 61.13 does not address this issue.
• Generally, facilities are required to notify EPA and the delegated agency of the delay. In some
instances, however, facilities are only required to notify the delegated agency of the delay.
Notification to EPA in addition to the delegated agency is dependent on individual Regional
delegations of these requirements. Written notification should be sent to the appropriate
Page 11
state/local agency and, if required, concurrently to EPA. The rescheduled test date should be
acceptable to both the delegated agency and the facility. This affords the delegated agency an
opportunity to observe the test, if desired. If timely notification is not provided, the test results
may be deemed unacceptable, and the source may be required to test again.
• For stack tests that are being conducted pursuant to requirements in an operating permit or an
enforcement order, the time frame for notification may differ and will be governed by the permit
or order.
• Notification is not necessary if the stack test is not within the scope of this guidance as
discussed in the Section, "Scope of Guidance." However, facilities should notify EPA and the
delegated agency if there is a potential for applicable limits to be exceeded. Furthermore, as
noted previously, the data from stack tests may be subject to Title V reporting requirements and
need to be considered by the source when submitting reports and certifying compliance pursuant
to the Title V program.
• 40 CFR Parts 60 and 61 do not require facilities to submit site-specific test plans prior to
conducting a stack test. 40 CFR § 63.7(b)(1) requires submission of such plans "upon request."
See also 40 CFR § 63.7(c)(2)(i) (owner or operator shall submit site-specific test plan if
requested by the Administrator). However, many delegated agencies routinely request that the
plans be submitted at the time of notification for review and approval. The submission of a plan
prior to the stack test helps to ensure that the testing requirements are interpreted correctly and
required test methods are followed; minimizes potential problems encountered during the test;
and reduces the possibility of testing errors. Ultimately, having the plan reviewed and approved
prior to the test reduces the number of retests.
• The format of site-specific test plans may vary. However, certain basic elements should be
addressed in a site-specific test plan to assist in national consistency, and ensure that a complete
and representative stack test is conducted. 40 CFR § 63.7(c)(2)(i) states that before conducting a
required performance test, the owner or operator shall develop a site-specific test plan and, if
required by the Administrator, submit it for approval. The test plan shall include "a test program
summary, the test schedule, data quality objectives, and both an internal and external quality
assurance (QA) program." Data quality objectives are "the pretest expectations of precision,
accuracy, and completeness of data." 40 CFR § 63.7(c)(2)(i). The internal QA program shall
include, "at a minimum, the activities planned by routine operators and analysts to provide an
assessment of test data precision; an example of internal QA is the sampling and analysis of
replicate samples." § 63.7(c)(2)(ii). The external QA program shall include, "at a minimum,
application of plans for a test method performance audit (PA) during the performance test."
§ 63.7(c)(2)(iii). In addition, a site-specific test plan generally should include chain of custody
documentation from sample collection through laboratory analysis including transport, and
should recognize special sample transport, handling, and analysis instructions necessary for each
set of field samples. For a prototype of a sufficiently detailed site-specific test plan, see
Emission Measurement Center Guideline Document (GD-042), "Preparation and Review of
Site-Specific Emission Test Plans," (March 1999) (www.epa.gov/ttn/emc/guidlnd.html).
Page 12
• To assist in the preparation and transcription of test plans, the Electronic Reporting Tool
(ERT) should be used when possible. (www.epa.gov/ttn/chief/ert/ert_tool.html). The ERT was
designed to replace the time-intensive manual preparation and transcription of stationary source
emissions test plans and reports currently performed by contractors for emissions sources, and
the time-intensive manual quality assurance evaluations and documentation performed by the
Regions or state/local agencies. The ERT provides a format that:
- Highlights the need to document the key information and procedures required by the
existing EPA Federal Test Methods.
- Facilitates coordination among the source, the test contractor, and the regulatory agency
in planning and preparing for the emissions test.
- Provides for consistent criteria to quantitatively characterize the quality of the data
collected during the emissions test.
-Standardizes the reports.
- Provides for future capabilities to electronically exchange information in the reports
with facility, state or Federal data systems.
• Test plans should be maintained by the facility consistent with the statutory and regulatory
requirements, and made available to EPA, and state/local agencies upon request.
• If a facility wishes to deviate from a required test method, the facility would need to gain
approval from the delegated agency in advance of the test. See 40 CFR § 60.8(b) (NSPS);
40 CFR § 61.13(h)(1) (NESHAP); 40 CFR § 63.7(e)(2) (MACT). For purposes of the NSPS and
NESHAP programs, changes are divided into two separate categories: "minor " changes; and
"major " changes (described in the regulations as alternative or equivalent methods). Major
changes must be approved by OAQPS, while minor changes can be delegated to state/local
agencies. See Memoranda from Jack R. Farmer to Allyn M. Davis, "Delegation of New Source
Performance Standards Authority to States " (February 24, 1983); and from Jack R. Farmer to
David P. Howekamp, "Delegation of NESHAP Authority to State/Local Agencies "
(December 17, 1984), both included in Attachment 2 to the guidance document entitled "How to
Review and Issue Clean Air Act Applicability Determinations and Alternative Monitoring,"
EPA 305-B-99-004, (February 1999). For examples of what constitutes major versus minor
changes, see the above cited memoranda.
• For purposes of the MACT program, changes to test methods are divided into three categories:
"major," "intermediate," and "minor ". Major changes must be approved by OAQPS, while
intermediate and minor changes can be delegated to state/local agencies. See 40 CFR §
63.91(g). Definitions of the three categories are provided in 40 CFR § 63.90.
• The facility must receive prior written approval for deviations from a test method from the
Page 13
appropriate delegated agency. If the deviation is to be approved by a state/local agency, it
should be in consultation with EPA, or as otherwise required by the delegation. See also "How
to Review and Issue Clean Air Act Applicability Determinations and Alternative Monitoring,"
EPA 305-B-99-004, Section 4.2, pp.19-22 (February 1999). If a deviation from a test method
has not been approved, the test results may be deemed unacceptable, and the source may be
required to test again.
• The request for a minor change or deviation from a required test method may be submitted as
part of the site-specific test plan, while intermediate and major changes or deviations to test
methods should be requested via written correspondence to the delegated agency or EPA as
appropriate. Requests for all changes or deviations must document to the satisfaction of the
delegated agency the requested change, and the rationale for the change. For a more detailed
guideline regarding the content for requests for changes to test methods, see Emission
Measurement Center Guideline Document (GD-022r3), "Handling Requests for Approval of
Minor/Major Modifications/Alternatives to Testing and Monitoring Methods or Procedures "
at http://www.epa.gov/ttn/emc/guidlnd.html.
• In addition to any deviations from the required test methods, the facility should document
within the test plan any adjustments that will be made prior to the stack test such as tuning the
burner or changing bags in a baghouse. It is not necessary, however, to describe normally
scheduled periodic maintenance that may occur in the normal course of operation and
maintenance of a unit. If an agency representative is present to observe the test, the facility also
should notify the observer of such adjustments before the test begins.
• Stack tests to determine and demonstrate initial compliance may be, in some instances, the
only test an emission unit will receive for an extended period of time. Therefore, all units should
be tested for initial compliance unless a waiver has been granted by the delegated agency
pursuant to 40 CFR §§ 60.8(b)(4), 61.13(h)(1)(iii), or 63.7(h). Waivers are granted only if the
owner or operator of a source has demonstrated by other means that the source is in compliance
with the applicable standard, or, under the MACT provisions, if the source is operating under an
extension of compliance pursuant to § 63.6(i), or has requested such an extension and the request
is under consideration by the delegated agency. The waiver regulations make clear that the
burden of proof is on the affected facility to justify the need for a waiver. Although the NSPS
and NESHAP programs do not specify what information is required as justification, the MACT
program in 40 CFR § 63.7(h)(3)(iii) states that the waiver application should include information
such as the "technical or economic infeasibility, or the impracticality, of the affected source
performing the required test."
• The primary issue of concern with respect to waiver requests is whether stack tests to
determine and demonstrate on-going compliance with emission limits should be waived under
the NSPS, NESHAP and MACT programs for units identical to a unit(s) that has been tested.
• Units, although identical in terms of design and control devices, may have process operations
that significantly alter their performance and ability to comply with the underlying regulatory
requirements on a continuing basis. Therefore, if the identical units have the ability to emit a
pollutant in excess of the prescribed emission limit, a stack test should not be waived without
adequate justification. However, a waiver may be appropriate on a case-by-case basis when
criteria such as the following are met:
(1) the units are located at the same facility;
(2) the units are produced by the same manufacturer, have the same model number or
other manufacturer’s designation in common, and have the same rated capacity and
operating specifications;
(3) the units are operated and maintained in a similar manner; and
(4) the delegated agency, based on documentation submitted by the facility,
(a) determines that the margin of compliance for the identical units tested is
significant and can be maintained on an on-going basis; or
(b) determines based on a review of sufficient emissions data that, though the
margin of compliance is not substantial, other factors allow for the determination
that the variability of emissions for identical tested units is low enough for
confidence that the untested unit will be in compliance.3 These factors may
include, but are not limited to, the following:
(i) historical records at the tested unit showing consistent/invariant
load;
(ii) fuel characteristics yielding low variability (e.g., oil)
and therefore assurance that emissions will be constant and
3 As a general matter, the greater the quantity of available emissions data, the smaller the
range of uncertainty about emissions and the more readily reviewing agencies can determine
precise levels of emissions variability. Under such circumstances, delegated agencies may have
greater assurance that compliance will be continuous even where the difference between actual
and permitted emission rates is relatively small.
Page 10
below allowable levels;
(iii) statistical analysis of a robust emissions data set
demonstrate sufficiently low variability to convey
assurance that the margin of compliance, though small, is
reliable.
• If a facility does not have the ability to emit a pollutant in excess of the prescribed emission
limit, waivers on a case-by-case basis may be issued for both initial and on-going compliance
stack tests. For example, a stack test waiver for identical units at a facility operating multiple
natural gas-fired boilers subject to a particulate matter standard generally would be appropriate.
• Waivers can be granted only by the appropriate delegated agency. See 40 CFR § 63.91(g).
See also, "How to Review and Issue Clean Air Act Applicability Determinations and
Alternative Monitoring," EPA 305-B-99-004, Section 4.2, pp.19-22 (February 1999). If the
delegated state/local agency has the authority to grant a waiver, it still should consult promptly
with EPA to promote national consistency.
3. STACK TEST NOTIFICATIONS
• The primary issue is what constitutes sufficient notification of a planned stack test under the
regulatory requirements. Sufficiency is defined to include both the timing of the notification, as
well as the content of the notification.
• Unless specified otherwise in the subpart, both the NSPS and NESHAP programs require at
least thirty (30) calendar days advance notice of a stack test [40 CFR § 60.8(d) and 40 CFR
§ 61.13(a) and ©], while the MACT program requires at least sixty (60) calendar days [40 CFR
§ 63.7(b)(1)]. The test date(s) and approximate start/end time of the test should be acceptable to
both the delegated agency and the facility to allow the delegated agency an opportunity to
observe the test, if desired. If for some reason the stack test must be delayed, facilities also are
required to provide notification of the delay. The time frame for such notifications differs under
each program. Under 40 CFR § 60.8(d), the facility is required to provide notification "as soon
as possible of any delay in the original test date, either by providing at least 7 days prior notice
of the rescheduled date of the performance test, or by arranging a rescheduled date with the
Administrator (or delegated state or local agency) by mutual agreement." Under 40 CFR
§ 63.7(b)(2), if the facility must delay the test due to "unforeseeable circumstances beyond [its]
control ", the facility must notify the "Administrator as soon as practicable and without delay
prior to the scheduled performance test date and specify the date when the performance test is
rescheduled." 40 CFR § 61.13 does not address this issue.
• Generally, facilities are required to notify EPA and the delegated agency of the delay. In some
instances, however, facilities are only required to notify the delegated agency of the delay.
Notification to EPA in addition to the delegated agency is dependent on individual Regional
delegations of these requirements. Written notification should be sent to the appropriate
Page 11
state/local agency and, if required, concurrently to EPA. The rescheduled test date should be
acceptable to both the delegated agency and the facility. This affords the delegated agency an
opportunity to observe the test, if desired. If timely notification is not provided, the test results
may be deemed unacceptable, and the source may be required to test again.
• For stack tests that are being conducted pursuant to requirements in an operating permit or an
enforcement order, the time frame for notification may differ and will be governed by the permit
or order.
• Notification is not necessary if the stack test is not within the scope of this guidance as
discussed in the Section, "Scope of Guidance." However, facilities should notify EPA and the
delegated agency if there is a potential for applicable limits to be exceeded. Furthermore, as
noted previously, the data from stack tests may be subject to Title V reporting requirements and
need to be considered by the source when submitting reports and certifying compliance pursuant
to the Title V program.
• 40 CFR Parts 60 and 61 do not require facilities to submit site-specific test plans prior to
conducting a stack test. 40 CFR § 63.7(b)(1) requires submission of such plans "upon request."
See also 40 CFR § 63.7(c)(2)(i) (owner or operator shall submit site-specific test plan if
requested by the Administrator). However, many delegated agencies routinely request that the
plans be submitted at the time of notification for review and approval. The submission of a plan
prior to the stack test helps to ensure that the testing requirements are interpreted correctly and
required test methods are followed; minimizes potential problems encountered during the test;
and reduces the possibility of testing errors. Ultimately, having the plan reviewed and approved
prior to the test reduces the number of retests.
• The format of site-specific test plans may vary. However, certain basic elements should be
addressed in a site-specific test plan to assist in national consistency, and ensure that a complete
and representative stack test is conducted. 40 CFR § 63.7(c)(2)(i) states that before conducting a
required performance test, the owner or operator shall develop a site-specific test plan and, if
required by the Administrator, submit it for approval. The test plan shall include "a test program
summary, the test schedule, data quality objectives, and both an internal and external quality
assurance (QA) program." Data quality objectives are "the pretest expectations of precision,
accuracy, and completeness of data." 40 CFR § 63.7(c)(2)(i). The internal QA program shall
include, "at a minimum, the activities planned by routine operators and analysts to provide an
assessment of test data precision; an example of internal QA is the sampling and analysis of
replicate samples." § 63.7(c)(2)(ii). The external QA program shall include, "at a minimum,
application of plans for a test method performance audit (PA) during the performance test."
§ 63.7(c)(2)(iii). In addition, a site-specific test plan generally should include chain of custody
documentation from sample collection through laboratory analysis including transport, and
should recognize special sample transport, handling, and analysis instructions necessary for each
set of field samples. For a prototype of a sufficiently detailed site-specific test plan, see
Emission Measurement Center Guideline Document (GD-042), "Preparation and Review of
Site-Specific Emission Test Plans," (March 1999) (www.epa.gov/ttn/emc/guidlnd.html).
Page 12
• To assist in the preparation and transcription of test plans, the Electronic Reporting Tool
(ERT) should be used when possible. (www.epa.gov/ttn/chief/ert/ert_tool.html). The ERT was
designed to replace the time-intensive manual preparation and transcription of stationary source
emissions test plans and reports currently performed by contractors for emissions sources, and
the time-intensive manual quality assurance evaluations and documentation performed by the
Regions or state/local agencies. The ERT provides a format that:
- Highlights the need to document the key information and procedures required by the
existing EPA Federal Test Methods.
- Facilitates coordination among the source, the test contractor, and the regulatory agency
in planning and preparing for the emissions test.
- Provides for consistent criteria to quantitatively characterize the quality of the data
collected during the emissions test.
-Standardizes the reports.
- Provides for future capabilities to electronically exchange information in the reports
with facility, state or Federal data systems.
• Test plans should be maintained by the facility consistent with the statutory and regulatory
requirements, and made available to EPA, and state/local agencies upon request.
• If a facility wishes to deviate from a required test method, the facility would need to gain
approval from the delegated agency in advance of the test. See 40 CFR § 60.8(b) (NSPS);
40 CFR § 61.13(h)(1) (NESHAP); 40 CFR § 63.7(e)(2) (MACT). For purposes of the NSPS and
NESHAP programs, changes are divided into two separate categories: "minor " changes; and
"major " changes (described in the regulations as alternative or equivalent methods). Major
changes must be approved by OAQPS, while minor changes can be delegated to state/local
agencies. See Memoranda from Jack R. Farmer to Allyn M. Davis, "Delegation of New Source
Performance Standards Authority to States " (February 24, 1983); and from Jack R. Farmer to
David P. Howekamp, "Delegation of NESHAP Authority to State/Local Agencies "
(December 17, 1984), both included in Attachment 2 to the guidance document entitled "How to
Review and Issue Clean Air Act Applicability Determinations and Alternative Monitoring,"
EPA 305-B-99-004, (February 1999). For examples of what constitutes major versus minor
changes, see the above cited memoranda.
• For purposes of the MACT program, changes to test methods are divided into three categories:
"major," "intermediate," and "minor ". Major changes must be approved by OAQPS, while
intermediate and minor changes can be delegated to state/local agencies. See 40 CFR §
63.91(g). Definitions of the three categories are provided in 40 CFR § 63.90.
• The facility must receive prior written approval for deviations from a test method from the
Page 13
appropriate delegated agency. If the deviation is to be approved by a state/local agency, it
should be in consultation with EPA, or as otherwise required by the delegation. See also "How
to Review and Issue Clean Air Act Applicability Determinations and Alternative Monitoring,"
EPA 305-B-99-004, Section 4.2, pp.19-22 (February 1999). If a deviation from a test method
has not been approved, the test results may be deemed unacceptable, and the source may be
required to test again.
• The request for a minor change or deviation from a required test method may be submitted as
part of the site-specific test plan, while intermediate and major changes or deviations to test
methods should be requested via written correspondence to the delegated agency or EPA as
appropriate. Requests for all changes or deviations must document to the satisfaction of the
delegated agency the requested change, and the rationale for the change. For a more detailed
guideline regarding the content for requests for changes to test methods, see Emission
Measurement Center Guideline Document (GD-022r3), "Handling Requests for Approval of
Minor/Major Modifications/Alternatives to Testing and Monitoring Methods or Procedures "
at http://www.epa.gov/ttn/emc/guidlnd.html.
• In addition to any deviations from the required test methods, the facility should document
within the test plan any adjustments that will be made prior to the stack test such as tuning the
burner or changing bags in a baghouse. It is not necessary, however, to describe normally
scheduled periodic maintenance that may occur in the normal course of operation and
maintenance of a unit. If an agency representative is present to observe the test, the facility also
should notify the observer of such adjustments before the test begins.
 |
| |
 |
| |
 |
| |
 |
| |
 |
| |
 |
| |
 |
