Active Leak Detection Program
The City of Kawartha Lakes has completed a Water and Energy Conservation Study, which included a review of water loss. As part of the City’s ongoing water loss management program, the City has developed and implemented an active leak detection program using District Meter Areas (DMAs). The AWWA M36 Manual “Water Audits and Loss Control Programs” recommends using DMAs for active leak detection, which is considered to be a North American Best Management Practice (BMP).
The Water Loss Mitigation Study recommended that District Meter Area (DMA), active leak detection, be implemented throughout the City. Specific focus of the DMA’s should be prioritized to growth areas and those distribution systems identified in this report with high water loss. There are 21 water distribution systems, of which many are relatively small. However, the Town of Lindsay has a large complex water distribution system, and there are also medium sized distribution systems at Bobcaygeon and Fenelon Falls. The highest percentage of water loss was estimated to be in Bobcaygeon, and the highest volume of water loss was estimated to be in Lindsay. Also, the other significant volume of water loss was estimated to be in Fenelon Falls. These three areas of Lindsay, Bobcaygeon and Fenelon Falls have also been identified as areas of growth in the City. In 2012 the City commenced implementation of DMAs in Bobcaygeon, Fenelon Falls and the Town of Lindsay to address significant water loss so water savings can be achieved.
District Meter Area Methodology
The DMA methodology forms the core of the AWWA Manual, M36, “Water Audits and Loss Control Programs” which was published in May, 2009. In the M36 manual, this flow based method of identifying areas with leakage is called “District Meter Areas – DMAs”. The M36 manual includes the most updated methodology and equipment. DMAs are considered to be a North American Best Management Practice (BMP) for water loss reduction.
The leakage management program ensures the losses are in fact leakage and not unidentified use or other data problems, such as meter error. The leakage management program also identifies where leakage is occurring, so that it can be located and repaired efficiently. One of the most reliable and long established methodologies to help with these problems is to split the distribution system into small sectors often referred to as District Meter Areas (DMAs).
Where possible DMAs are designed so they can be fed through a metered water feed throughout the day and night. Once these areas have been established it is then possible to use the data on flow and pressures collected throughout the 24 hour period, knowing the likely makeup of consumption, and estimate losses for the area.
The City has currently implemented DMAs and Intensive Acoustic Leak Detection in the following areas:
- DMA in Fenelon Falls Drinking Water System (DWS)
- DMA in Bobcaygeon Drinking Water System (DWS)
- Intensive acoustic leak detection in the downtown area of Lindsay
- Two DMAs in the Town of Lindsay (scheduled for completion in 2014)
Night flow measurements including step testing, sounding and correlating have been completed at night when ambient noise from busy road traffic or high water demand is minimal. The Division has calculated Legitimate Night Flow (LNF) by installing insertion meters at Fenelon Falls and Bobcaygeon DWS. Night flows were measured by the DMA flow meter installed at the water storage tank outlet main in Bobcaygeon and Fenelon Falls.
The measured flow data estimated potential leakage where designs of step-test areas were completed followed by sounding and correlation to pinpoint potential leaks. After the leaks were located and repaired, night flow data was measured and reviewed again to establish if more potential leakage existed. To date the program has found and repaired a number of leaks equating to approximately 30 m3/hr of reduced water loss.
Lindsay DWS Intensive Acoustic Leak Detection
In order to maximise the detection of leaks using acoustic methods, intensive leak detection was performed in the downtown area of Lindsay. The majority of leak detection work was carried out at night to assist in the successful detection of leaks. This activity included sounding of all hydrants and valves along with sounding of curb stops where a ‘Leak Noise Map’ of the area was developed.
Sounding utilizes sensitive acoustic listening devices called Ground Microphones. The noise produced from pressurized water forcing its way through a crack or joint makes a distinct sound that can be heard on valves, pipes or services when listened to. The traditional sounding technique uses a wooden or metal rod to listen for leak which is known as a ‘listening stick’. This technique is a reasonably low cost operation with a moderately high success rate on metallic pipes only. The leak detection engineer will place their listening stick on main fittings such as valves and hydrants which allows them to listen for the sound of escaping water from the water distribution system. The leak noise is then transmitted from the fitting to the engineer’s trained ear via the listening stick, similar to a doctor listening to a heartbeat through a stetho- scope.
Leak Detection was then focused on areas that indicated potential leakage using the ‘Leak Noise Map’ developed through sounding. Potential leak sites were pinpointed further by Leak noise correlation testing. The Correlation test works by measuring pipe noise at two locations that pinpoint the position of a suspected leak. These tests are carried out using electronic instrumentation that measures leak noise at two points along the water main. The turbulence caused by the leak creates a noise which is picked up by the sound waves passed through the pipe from point to point were the position of the leak is determined based on the time shift between the two signals.
Principles of Water Loss Management
A major initiative to understand leakage was carried out in the United Kingdom in the early 1990’s. As a result of this and subsequent research work, the factors behind the processes leading to real losses are now well understood. The most significant step taken to develop this understanding was to break leakage into a number of components and then look at the factors that influenced the losses from each of these components.
Leakage was broken down into a two dimensional matrix of components by type of leakage and by asset type as follows:
Leakage Types was split into three categories as follows:
- Background Leakage
- Leakage from Reported Breaks/Leaks
- Leakage from Unreported Breaks/Leaks
Assets types are categorized as follows:-
- Trunk Mains/Service Reservoirs
- Distribution Mains
- Service pipes to the edge of street
- Service pipes from edge of street to meter
Background leakage is defined as leakage below the level of detection (by currently available techniques). It is primarily composed of weeping joints and gaskets rather than holes in pipes. Analysis of minimum levels of leakage detected in small areas established the expected losses from background leakage based on the length of mains and number of connections. It also found that these losses were heavily related to pressure; such that if pressure is reduced by 10% then background leakage would reduce by 15%. Since background leakage is, by definition, below the level of detection, pressure management is the only way of reducing losses from this type of leakage other than replacement of the assets with new joints.
Leakage from Reported Breaks/Leaks
Reported breaks/leaks are those that come to the attention of the City operator from customer reports. Leaks are detected when water is surfacing or is causing supply problems such as low pressure or no water. Water is lost until the leak is repaired or isolated. The volume of water lost is proportional to the frequency at which leaks break out, the proportion that are reported, the length of time the leak runs and the flow rate. The length of time a leak runs is dependent on the utility policy on inspection and repair.
Leakage from Unreported Breaks/Leaks
A leak that either comes to the surface or causes a supply problem typically will not come to the attention of the City. In this case these leaks can accumulate and the losses from leakage will continue to rise over time. In order to control this rise in leakage (often referred to as the Natural Rate of Rise of Leakage or NRR) it is necessary to proactively look for these leaks. This would normally be carried out in response to a rise in the output at treatment works/pumping stations etc. or to rises in flows being measured at meters within the Distribution system. The volume of water lost will again be proportional to the frequency at which leaks break out, the proportion that go unreported, the length of time a leak runs and the flow rate.
The flow rate is proportional to pressure in the same way as the flow rate of reported leaks. The length of time a leak runs is dependent on municipality’s policy on leakage detection and repair. In this case a leak can run for a very long time. For example, if the municipality policy is to complete a leakage sounding survey on an annual basis, then a leak will run for 180 days on average before being detected. If there is no policy on leakage detection then leaks may have been running for many years. The process of managing leakage from unreported leaks is often referred to proactive leakage control, or active leak detection – as the operator has to be proactive in identifying that unreported leaks are accumulating on the network
The main influences and therefore the management of leakage are illustrated in the figure below. In the diagram the size of the box in the centre represents the volume of water lost from leakage. The four arrows show that the size of the box (leakage) can be influenced by changing these main items, namely:-
Pressure Management – reducing pressure will reduce the level of background leakage, the rate at which water is lost from leaks and the frequency at which leaks break out.
Speed and Quality of Repair – reducing the time taken to repair known leaks will reduce the quantity of water lost from leaks. The quality at which these repairs are carried out will ensure that leaks do not break out again
Speed of Detection – the water lost from unreported leaks is heavily dependent on the policy on proactive leakage control
Infrastructure Renewal – renewing the assets should reduce the frequency at which leaks break out
Leakage management can be defined as establishing the feasibility and cost effectiveness of “squeezing” the box.