Enhanced Leak Detection

Bader, Laura and Billingham, John and Callahan, Daniel and Catrina, Florin and Edwards, David and Fehribach, Joseph and Gemmrich, Simon and Hazaveh, Kamyar and Johnson, Katharine and Moore, Richard and Mykrantz, Andrew and Ni, Peng and Phillips, Joel and Tilley, Burt and Ware, Kimberly and Weekes, Suzanne (2005) Enhanced Leak Detection. US Workshop on Mathematical Problems in Industry > 21st MPI [Worcester 13/6/2005 - 17/6/2005].

Full text available as:

	PDF (Study group Report) - Requires Adobe Acrobat Reader or other PDF viewer.
	PDF (Patent) - Requires Adobe Acrobat Reader or other PDF viewer.

Abstract/Summary

A key requirement for Veeder-Root’s Enhanced Leak Detection System is that it be able to test in situ for the presence of leaks at gasoline dispensing facilities. Aside from the obvious issues of safety and lost product, this functionality is obligatory for compliance with environmental standards mandated by federal and state oversight bodies, such as the California State Water Resources Control Board (SWRCB). The SWRCB demands a testing procedure that includes conditions as close to operational as possible, while still using environmentally safe gases as a test fluid. Although the test parameters (e.g., pressure) are allowed to deviate from operating conditions in order to facilitate the test procedure, a prescribed rescaling of the test thresholds must then be applied to account for the deviation. Whether the test is run at operation conditions or in a slightly different parameter regime, the fact that the testing must be done on the product and return lines after installation at a service station presents significant challenges in devising an effective test strategy.

Problem Statement

The California State Water Resources Control Board (SWRCB) requires new gasoline station fuel containment and dispensing systems, before being put into operation, to pass a very stringent leak detection test. A successful detection system must detect a leak of 0.005 gallons per hour of a test fluid (gas or liquid) at least 95% of the time and have a false alarm rate of no more that 5% for a non-leaking system. Currently this requirement is met by only one vendor that uses an organic tracer gas and air sampling technique to find very small leaks. This solution is very expensive for gasoline station owners.

Veeder-Root is interested in exploring less expensive methods that employ parts or extensions of our existing fuel and vapor monitoring and leak prevention equipment and techniques. Solutions are aimed at double-wall contained underground storage tanks (UST's), pipes, and containment sumps that have inner (primary) and outer (secondary) containment walls with an interstitial secondary containment space between the walls.

In a working storage and fueling system with Veeder-Root monitoring equipment, a submersible turbine pump (STP) in each UST provides the pressure and flow needed to supply fuel to the dispensers. A vacuum is applied by the automated leak prevention and monitoring system to the interstitial space between primary and secondary walls in a multi-zoned fashion. Typically each pipe, tank, and containment sump is controlled as a separate vacuum zone via solenoid operated valves and vacuum sensors. The monitor controls the vacuum levels with the valves and a common vacuum source while using sensors to monitor the fuel line pressures and interstitial space vacuum levels as needed and where needed. As long as vacuum levels can be maintained, any primary wall leak is contained within the interstitial space and prevented from reaching the environment. If a substantial loss of vacuum is detected, it is indicative of a leakage problem. A liquid trap with sensor is provided in each zone. Liquid fuel that reaches a trap is detected and is also indicative of a leakage problem. See, for example, U.S. Patent Nos. 6,834,534, and U.S. Patent Application Publication No. 2005/0039518 A1.

An ELD test should be performed prior to filling the new containment system with fuel. The challenge is to design a control and monitoring scheme that uses automatically controlled pressure and/or vacuum levels and associated sensors to determine 1) if there is a leak of at least 0.005 gph, 2) certain characteristics of the leak (to be discussed in more detail at the problem session meeting), and if possible in the case of piping, 3) where the leak is located along the length of the pipe.

This will require some understanding of the nature of inert gas (e.g. nitrogen) and/or air flow through leakage paths entering into relatively large volume interstitial spaces and it's effects on vacuum or pressure levels under conditions of un-measurable thermal or other interfering effects.

A successful outcome from the 1-week workshop will provide an automated 0.005 gph leak detection test logic scheme or schemes, associated mathematical relationships, and an assessment of practical feasibility using assumed test conditions, parameters, and interfering factors. The test scheme must be achievable in a real-time microcomputer controlled leak monitoring system with practical computing power and speed.

Archive Staff Only: edit this record