Leak detection and repair (LDAR) programs are required as part of the standards established in 40 Code of Federal Regulations (CFR) 60 (NSPS), 40 CFR 61 (NESHAP), 40 CFR 63 (MACT), and 40 CFR 264 (Hazardous Waste Handling). An LDAR program is a facility’s system of procedures used to locate and repair leaking components (e.g., valves, pumps, connectors, compressors, and agitators) to minimize fugitive volatile organic compounds (VOC) and hazardous air pollutants (HAP) emissions. To verify minimization of VOC and HAP emissions, the U.S. EPA has been conducting audits and pursuing enforcement actions in the petroleum refining and chemical manufacturing industries since the late 1990’s. Both large and small facilities have been reviewed under this coordinated U.S. EPA program.
EPA’s Audit Program
Initially, EPA focused on petroleum refineries for its LDAR-related enforcement activities. Since 2005 and continuing into 2010, EPA has also concentrated enforcement efforts in the chemical industry (e.g., facilities subject to Hazardous Organic NESHAP and Pharmaceutical MACT standards).1,2 EPA, and in particular Region V, has announced that it will focus on Miscellaneous Organic NESHAP (MON) subject facilities in the forthcoming years.2 As part of the most recent LDAR enforcement initiative, Region V has completed 38 investigations, issued 22 violation notices, referred 9 cases to the Department of Justice, and settled 12.3 As shown in recent notices of violations (NOVs) and/or findings of violations (FOVs), the compliance issues discovered by EPA at chemical manufacturing facilities encompass all facets of the LDAR standards.
Common EPA Audit Findings: Method 21 Issues
For both petroleum refineries and chemical manufacturing facilities, EPA has cited inappropriate application of 40 CFR 60, Appendix A, Method 21 (Method 21). Method 21 requires a portable monitoring device for identifying components emitting VOC/HAP above the leak definitions (e.g., 10,000 ppmv or 500 ppmv). One of the primary methods by which the U.S. EPA assesses adherence to the Method 21 is “comparative monitoring.” Often when conducting comparative monitoring, the EPA auditors will observe the facility LDAR technician while conducting parallel monitoring per Method 21, which includes observing all of the following facets of Method 21:
- Zero and leak definition calibration checks,
- Calibration adjustments,
- Instrument response time checks,
- Time spent monitoring per component, and
- Leak interface monitoring techniques for each component.
Upon completion of the monitoring, the EPA will compare its leak rates (e.g., percentage of leaking components to number of components monitored) with the facility technician’s leak rates. On average, the leak rate determined by EPA is about 5% while the leak rate determined by facility technicians is about 1%.2 Discrepancies between the leak rates can be due to any one of the above listed factors and increases the level of scrutiny with respect to facility leak rates previously reported. For example, according to EPA, being one centimeter away from the interface can cause the technician to miss a defined leak.3
Common EPA Audit Findings: Periodic Monitoring Issues
Method 21 is utilized at subject facilities to periodically monitor each regulated component. The frequency of such monitoring may vary from monthly to every eight years depending on the subpart and the component being monitored. If the component is found to be leaking, it is tagged and must be repaired within a specified period of time.
Please note that these are leak detection programs; therefore, leaks are expected to be discovered. Finding a leak is not always a violation; however, not conducting appropriate monitoring, recordkeeping, reporting, and repairs would be violations. Therefore, review of recordkeeping (e.g., repair timing records and leak rate calculations), reporting (e.g., delay of repair), and monitoring frequencies are key components to any EPA inspection. For example, violations listed in EPA NOVs and/or FOVs include the following:
- Not conducting monitoring of each regulated component every monitoring period
- Classifying insulated valves as exempt from Method 21 monitoring,4
- Not maintaining required lists (e.g., difficult-to-monitor (DTM), unsafe-to-monitor (UTM), instrumentation systems, and pressure relief valves equipped with rupture disks)
- Not reporting delay of repair technical infeasibility information in the periodic reports, and
- Not maintaining complete leak repair records.
Common EPA Audit Findings: Leak Repair Issues
Once a defined leak is discovered, facilities are required to clearly identify the component as leaking (e.g., leak tag). The leak tag must remain on the component until repaired. For certain components such as valves in gas/vapor service or in light liquid service, the leak tag must remain on the component until the follow-up monitoring (i.e., monitoring conducted within three months of repair) is completed.5
EPA has also noted failure to conduct the first attempt at repair within five days. First attempt at repair is generally defined in the MACT standards as follows:
To take action for the purpose of stopping or reducing leakage of organic material to the atmosphere, followed by monitoring as specified in §63.180(b) and (c), as appropriate, to verify whether the leak is repaired, unless the owner or operator determines by other means that the leak is not repaired.6
EPA expects an action be undertaken to attempt to reduce the instrument reading below the leak definition. Two examples of first attempts at repair for leaking pumps include:
- Tightening of packing gland nuts, and
- Ensuring that the seal flush is operating at design pressure and temperature.
If the first attempt at repair does not reduce the instrument reading below the leak definition, the repair must be completed within 15 days of leak discovery, unless the delay of repair provisions can be applied. The definition of repair in the MACT standards, unless otherwise noted in the applicable section, includes conducting Method 21 to confirm the physical repair with an instrument reading below the leak definition. Therefore, the Method 21 reading confirming the physical repair must be completed within 15 days of leak discovery. EPA has cited facilities for not completing timely repairs in situations where physical repair was completed within 15 days, but confirmation monitoring was completed after the 15 day period.
As stated above, Method 21 monitoring must occur within the 15 day window for a component to be considered “repaired.” However, if a potential leak is detected visually and confirmed to be a Method 21 leak, some LDAR regulations (e.g., MACT UU, Pharma MACT, NSPS VV, and NSPS VVa, HON “heavy liquid” components) allow for the potential leak to be considered repaired upon elimination of the visual indications.7 Examples where Method 21 must always be conducted to confirm repairs of visual leaks include, but not limited to the following:
- Pumps in light liquid service subject to the HON,
- Agitators in light liquid service subject to the HON, and
- Agitators in gas/vapor service subject to the HON.
Facilities must be familiar with the LDAR requirements of their specific regulations. While the LDAR regulations all have similar requirements, unrecognized nuances in each regulation can result in potential violations.
Common EPA Audit Findings: Open-Ended Lines
Open-ended lines are another common issue and one of the easiest issues for EPA to find. Per 40 CFR 63.161, as well as other definition sections of LDAR regulations,
Open-ended valve or line means any valve, except pressure relief valves, having one side of the valve seat in contact with process fluid and one side open to atmosphere, either directly or through open piping.
The only times facilities are allowed to have open-ended lines is when they are in use or due to safety concerns (e.g., containing materials that would autocatalytically polymerize, or would present an explosion, serious overpressure, or other safety hazard if capped or equipped with a double block and bleed system).
Inspectors can easily identify this issue during audit walkthroughs since no Method 21 monitoring is required, just the human eye. Simply not having a plug, blind flange, cap, or a second valve is a violation and therefore, easily spotted by inspectors. Even if the open-ended line is sealed immediately upon discovery by the inspector, it would still be an open-ended line. Therefore, a potential violation could be cited.
Inspectors commonly find open-ended lines at sampling connection systems. Sampling connection systems have their own LDAR requirements regarding maintenance of closed-purge/closed-loop systems. Additionally, the sampling connection systems must be configured and/or operated in a manner that does not result in an open-ended line. For instance, the two final valves should be at least within arms’ reach to allow facility personnel to close the valve contacting the process fluid and then close the second valve.
Common EPA Audit Findings: Component Identification/Percent Leaker Calculations
Facilities are required to identify LDAR-subject components. While physically tagging subject components is not required, components must be distinguishable from those not subject to LDAR regulations. The result is an accurate count of LDAR components. Not having an accurate component count affects New Source Review (NSR) applicability for construction/modification projects as well as percent leaker calculations. For example, the percent leaker calculations can be skewed if non-LDAR subject equipment (e.g., equipment in vacuum service) is included in the program.
Each standard specifies the equations to be used for the percent leaker calculations for each component type. The percent leaker calculations are utilized to determine monitoring frequency and if a quality improvement program is triggered. The percent leaker calculations under the MACT standards are not intended to include equipment that is designated as:
- UTM or DTM even if UTM and DTM components are monitored during the monitoring period,
- Heavy liquid components, and
- Valves and connectors part of an instrumentation system.
EPA has cited facilities for not performing percent leaker calculations appropriately.
As a result of recent LDAR audits and the associated issues, EPA is initiating enhanced LDAR programs as part of the most recent consent decrees.8 The enhanced LDAR programs for chemical facilities are more stringent than the enhanced LDAR required for the refining industry. Starting in 2000, refineries were required by consent decrees to implement enhanced LDAR programs. Enhanced LDAR programs include many elements. Three main elements of the refinery enhanced LDAR are as follows:
- Leak definitions were lowered from NSPS VV levels (e.g., 10,000 ppm) to HON levels (e.g., 500 ppm),
- Initial attempts at repair were required at concentrations as low as 200 ppm (not a “leak” but a lowered action level), and
- Periodic internal/third party audits were required.9
The chemical industry consent decrees are also requiring enhanced LDAR standards, similar to refinery enhanced LDAR with variations as well as additional requirements. The chemical industry enhanced LDAR programs also lowered the leak definitions, required periodic audits, and required replacement or repacking at lower concentrations during maintenance turnarounds. Other elements of the chemical industry enhanced LDAR requirements include:
- Certain technologies are required for repair of leaking valves and connectors,
- Closure devices associated with open-ended lines must be monitored via Method 21 on a periodic basis,
- More stringent delay of repair provisions, and
- Frequent periodic monitoring of certain valves, connectors, pumps, and agitators.
In the next issue of Environmental Quarterly, we will explore the elements of enhanced LDAR in detail.
1 Loukeris, K. and Wilwerding, J. (2008) LDAR: Enforcement and Compliance. Presented at ISA’s 8th LDAR Symposium, San Antonio, Texas. 1 April.
2 MacDowell, W, Section Chief, Air Enforcement & Compliance Assurance Branch, USEPA Region V. (2009) U.S. EPA Region V Enforcement Priorities for 2009/2010. Presented at All-Ohio Fall 2009 Multi-Media Technical Conference, Columbus, Ohio. 8 October.
3 Leak Detection And Repair - A Best Practices Guide (http://www.epa.gov/compliance/resources/publications/assistance/ldarguide.pdf )
4 While there is an exemption from monitoring for insulated connectors, no such exemption exists for insulated valves.
5 Another example where the tag must remain longer than leak repair dates are connectors subject to 40 CFR 63.174(c)(1)(i). Please refer to your applicable LDAR regulations for other examples.
6 40 CFR 63.161
7 Please note the 15-day repair window is not applicable for potential leaks not confirmed to be Method 21 leaks. Please refer to specific LDAR regulation for repair timing for these cases.
8 Please note not all chemical facilities that received LDAR-related consent decrees have been required to initiate enhanced LDAR programs.
9 The frequency of third party audits is dependent on each individual consent decree. It could be annual, biennial, every (5) years, etc.