Leak Detection is accomplished in two phases. During the first phase, the entire system is surveyed for “leak sounds” on any available contact point (valves, fire hydrants, service connections, blow-offs). When a sound is heard, the location is noted as a potential leak site. Actually, any condition which interferes with the normal flow of water can produce vibrations similar to the vibrations caused by leaks. During the second phase, each location is further investigated. If necessary, a computerized leak correlator that works on sonic transmission (speed of sound) principles is used to pinpoint the exact location of the leak. The correlator eliminates the need for extensive hit-or-miss excavation, and the unnecessary destruction of expensive pavement. The District is currently using an AccuCorr Leak Correlator manufactured by Fluid Conservation Systems, Inc. Any utility system requesting leak detection services will be asked to commit a vehicle and an individual with a good working knowledge of the distribution system to accompany and assist the Urban Mobile Lab operator. The time it takes to complete leak detection of a public supply system is dependent upon a number of factors to include the size, accessibility, and condition of the system. Upon completion, the Urban Mobile Lab will provide a written report to the utility director.
Meter Testing. The Urban Mobile Lab offers testing of both well and service meters. Meter testing is an essential element of any water audit. If the accuracy of meters is not determined, the results of a water audit cannot be considered valid. Any utility system requesting leak detection services will be asked to commit an individual with a good working knowledge piping in proximity to the meter(s) being tested.
Well meters are tested with a Panametrics PT868. The PT868 is a non-intrusive transit-time test meter, which calculates flow velocity by determining the time difference for a sonic signal to travel against the direction of flow in a pipe versus time difference for a sonic signal to travel with the direction of flow. When properly programmed with pipe data, the PT868 calculates flow based upon the area of a cross section of the pipe and the flow velocity, and is accurate to within plus or minus two percent (2%). Flow calculated by the PT868 is compared to meter readings recorded during the test period (typically 10 minutes) and the accuracy of the well meter can then be determined. In order to insure maximum accuracy, the PT868 must be installed on a straight run of pipe, ideally with ten (10) diameters of unobstructed pipe upstream of the PT868, and five (5) diameters of unobstructed pipe downstream. In some cases, piping in the vicinity of the meter being tested may need to be excavated prior to the test.
Service meters are tested using an MUN-1 Utility Service Analyzer. The MUN-1 is a small, hand-held unit which can be readily calibrated. The test is conducted by connecting a hose to one of the customer’s hose bibs, and allowing exactly 10 gallons through the MUN-1. The customer’s meter is observed during the test to determine if an equivalent volume of water flowed through it. Typically, a representative sample (1%-5%) of service meters are randomly tested, and then the average accuracy is calculated. The average accuracy is then assumed for all service meters when making adjustments to consumption data during the documentation of a water audit.