After initial meetings about suspected leakage and NRW losses with Craig Needham and Piotr Kozeil from Yorkshire Water a trial was arranged to use the GUTERMANN AQUASCAN TM2 correlator. An initial location on an 18” Cast Iron trunk main was identified from Longwood Service Reservoir feeding four large DMAs. Other leak detection methods had previously been carried out with a point of interest which were to be compared and followed up with the AQUASCAN TM2.
The Cast Iron main had several hydrants where we could use hydrophones at approximately 900 metres apart; the area is built up with a mixture of domestic and industrial properties also several large trees along the route of the proposed correlation.
The AQUASCAN TM2 using hydrophones was set up at 900 meters apart using aerial extensions to avoid signal loss which can be caused through tall buildings and large trees between the sensors. The pipe information was input into the AQUASCAN TM2 and a correlation ran. This was a long distance and it was expected to allow the correlation to run for some time, after approximately 5 minutes a peak developed at 520 meters with a quality of 100%. We measured out the distance to find a supply feeding a mill running with the meter not registering the flow. Other points of interest have been identified using peak suppression on this length of main.
Investigation 1: Longwood CT29 – 18” CI
Longwood CT29 – 18” Cast Iron. Correlation 1-2 (900m, 18” main CI). Peak 520 m from Sensor A (sensor 1).
A good noise was found on the supply (possibly missing consumption estimation of 300m3/d). This had not been picked up on previous leak detection exercises.
CT29 Flow Profile, Missing Consumption (Longwood) 2.197l/s (254m3/d).
18” Main CI – Under Investigation. ICA job raised to check customer meter.
Investigation 2: BT22 Shipley – 125mm MDPE
BT22 Shipley – MDPE main 125mm, 68.7m correlation.
The correlation peak shows leak at 24.7 metres from Sensor A.
Step test shows 2l/s drop.
Investigation 3: John Street Thornton, 420m correlation on Mixed Material
Sensor A was placed on the red spot, Sensor B was placed on the yellow spot. Distance 421 metres.
The correlation showed a clear correlation peak, ca. 15 metres from Sensor A and ca. 406 metres from Sensor B:
John Street Thornton: 3.5 l/sec drop:
Yorkshire Water has carried out further correlations on the Longwood trunk main with GUTERMANN AQUASCAN TM2. Repairs and NRW are now rectified with savings on this main collectively around 900m3; other correlations on various locations have also resulted in leaks found and further savings. As a result of the broad trials, Yorkshire Water has formed several trunk main leak detection teams throughout their undertaking successfully using AQUASCAN TM2 correlators.