Comment submitted by Chris Patterson, InspectionLogic Corporation

FROM: Chris Patterson, InspectionLogic Corp., www.leakdas.com 1-502-244-4500, cpatterson@InspectionLogic.com <br><br> RE: Docket ID No. EPA-HQ-OAR-2003-0199 - Alternative Work Practice to Detect Leaks From Equipment <br> COMMENT: This comment pertains to "Part 60 - AMENDED," 60.18 paragraph (i)(4)(v), "RECORDKEEPING." <br> <br> Is it technologically feasible to store video of ALL components inspected using optical gas imaging? Yes, but... The point of this Comment is to raise awareness of the cost associated with doing so and to propose a clarification to the AWP's recordkeeping requirements that would alleviate burden while maintaining the data of highest interest. <br><br> A motivator for this AWP is to lower the cost of compliance. This AWP also strives to focus a plant's operational resources on finding and repairing the most significant leakers ("...0.13 percent of components contribute greater than 90 percent of controllable fugitive emissions..."). <br><br> Paragraph (i)(4)(v) states, "A video record must be used to document the leak survey results." The most obvious interpretation of this is that all images observed in the field by the operator of an optical gas imaging instrument are to be recorded much the same way as you would using a home video camera. An electronic video recording device brought into the field with (or as part of) the optical gas imaging instrument would store the images of all inspected components as a digital file that could later be transferred to a computer for storage. Stored video files would then be available for retrieval and viewing, presumably by an auditor wanting to verify that each and every component was monitored using this AWP. <br><br> If the intent of this AWP is for regulated entities to create and store a video record of ALL regulated components every 30 to 60 days, we would like to offer the following observations: <br><br> 1) Storage space for electronic files that plants consider highly valuable and mandated by regulations is expensive. Video files are very large. One optical gas imaging camera manufacturer told us that their LDAR video files presently consume 0.2 megabytes per second (mpeg4 file at 512x384 resolution). If the camera operator surveys an area and determines that it takes 2.5 seconds* to ascertain whether or not a component is leaking, that's half a megabyte per component. For a plant with 100,000 components, that's 50 gigabytes PER INSPECTION, which will be repeated 6 to 12 times per year. For regulations that presently stipulate the need to keep records for 5 years, a plant with 100,000 components will need to maintain secure computer file storage on the order of 3,000 gigabytes. This introduces a new cost for regulated entities to comply with EPA regulations. <br> We can speculate that technology will eventually reduce video file size. In the above example, reducing the file size by 50% means the plant will still be maintaining (and backing up) 1,500 gigabytes, which is still significant. <br> Auditing over a 1,000 gigabytes of video to discover possible violations will offer its own challenges, but this is not a likely concern for regulated entities. <br> <i>*This scenario calls to mind a question that does not appear to be answered by the AWP.</i> How long will an operator have to record the image of each component to determine whether or not it is leaking? If there is no minimum time, will a 5 second pan of the camera to sweep across a broad area containing 10's or 100's of components be considered sufficient for compliance and record keeping? Will the EPA require and/or oversee the camera manufacturer to establish a minimum imaging time per component as part of the "manufacturer's operating parameters" mentioned in (i)(3)? <br><br> 2) Plant safety typically dictates that electronic equipment carried into hazardous areas be certified as intrinsically safe. Thus far, intrinsically safe versions of electronic data collection devices used for LDAR have been commercially available because such devices are routinely used in the plant for other purposes. We haven't seen much development of new data collection technology specifically to serve the LDAR market because by itself, the LDAR market is very small. Companies who develop and sell data collection and data storage technology expect to sell hundreds-of-thousands of units. By expecting the camera operator to record and store 5,000 megabytes of video (assuming he or she can image 10,000 components in a day, which some people expect to do), we are requiring that person to take into hazardous areas a data storage device that is not presently intended for that purpose. Examples of data storage devices that are presently being used with optical gas imaging cameras include consumer- grade digital video recorders and home video cameras that can store either the optical gas imaging equipment's output or visible light images on removable storage cards. These kinds of devices are necessary to accommodate storage of large video files, but may not be permitted in many areas of the plant where inspections must be performed. Until there is development of innovative technology in the areas of data storage and portable power for flammable environments, use of the cameras may be restricted. <br> In any case, the time necessary to transfer large files between the field data collection device and the plant's computer will need to be factored into the labor time necessary to use this AWP. <br><br> 3) It is not explicit in the proposed rule, but each and every component must still have a unique identifier. Component ID maintenance will not go away. This information must be kept up-to-date and accurate for record keeping and reporting purposes. The plant's LDAR database must still maintain records for each component, including whether or not it passed or failed its optical gas imaging inspection. <br><br> 4) The heart of Leak Detection And Repair compliance lies in the "And Repair" aspect of each plant's LDAR program. Under this new AWP, how leakers are handled remains unchanged (with the exception that if a leak is found using optical gas imaging, a passing retest must be performed the same way). As mentioned more than once in the proposed rule's preamble, a small number of components account for most of the emissions. This AWP aims to target the leakers quickly and at a low cost of field operations. <br><br> Given the observations described above, has consideration been given to the use of optical gas imaging for inspecting all components, but limiting the requirements to make and maintain a video record to only those components found to be leaking? <br><br> With the ease of performing inspections using the camera, auditors should be able to quickly discern the effectiveness and compliance of a plant's LDAR program by performing their own random inspections rather than reviewing past video of non-leaking components. <br><br> This AWP is essentially an "Enhanced Visual Inspection" with the result of each inspection being a simple pass or fail. These Visual Inspections can be stored in the Leak-Detection-and-Repair database as Pass or Fail, and that record can include video files for failed inspections and subsequent retests. <br><br> By storing only the video of failing inspections and their subsequent repairs, the amount of storage space required will be dramatically reduced, and field data collection will be simpler and much faster. It would readily allow the plant to "show" compliance for their small population of leaking components. <br><br><br> One other question on a different topic; mass emissions calculations. When using this AWP, will regulated entities use the Leak\No Leak calculation with new emission factors? <br> -----------------------------------