WATER: Listening Closely

Birmingham Water Works uses sophisticated technology to locate leaks and reduce nonrevenue water despite shortages of qualified people

Birmingham Water Works (BWW) is a big water system in a drought-stricken region. Leaks are an obvious concern, and in 1999 the nonrevenue water percentage had crept up to an expensive and unacceptable 25 percent.

With 600,000 customers and more than 4,000 miles of mostly ductile iron line to maintain, and with limited manpower, the utility had limited options. “A manual leak survey of the entire system takes 30 months,” says Revenue Water Depart-ment manager Geoff Goodwin.

“Since our main service area is also our oldest gradient, that section alone took 14 months to inspect. In 1999, with the high nonrevenue rate, we realized we had to rethink our approach and find a way to leverage technology.”

BWW still has 10 employees who use listening devices to find leaks, but now technology enables the agency to inspect its main service area for leaks in just nine weeks. The agency relies on Permalog leak noise loggers from Datamatic Ltd. The devices, automated sensors installed at valves, are the most important component of a three-pronged approach to reduce nonrevenue water. The other two prongs are adding more district-metered areas and using a sophisticated pressure-management system.

The application of practical technology has reduced nonrevenue water to a more reasonable 14 percent, without a staffing increase. Goodwin says reducing dependence on manpower is key. “A lot of people my age are seeing the available work force get smaller,” he says. “Even compared to just 10 years ago, we’re having a hard time finding people to fix leaks. We’re turning to technology because we have to.”

Automated detection

Permalog loggers have several features that in Goodwin’s view made them just right for Birmingham. They’re small. Being magnetic, they attach easily to valves. And they can cover up to a 1,000 feet of metallic line. They “wake up” at night, when ambient noise is low, and listen for acoustic signatures of leaks.

By applying algorithms, the sensors maintain a low rate of false positives. “We do get false hits,” says Goodwin. “The Permalog units are sensitive, and other sounds will set them off. Of course, we want them to be sensitive. Our crews are getting very good at using listening equipment to verify a leak before we dig.”

If a leak is detected, the sensors send low-frequency radio signals that can only be picked up by a nearby patroller unit. The limited range allows the units to avoid FCC licensing, and they do not cause radio interference. “Basically, it’s a remote listening device,” says Goodwin. “It turns itself on at two o’clock in the morning, then cuts itself off and turns back on at three and at four o’clock. If it hears something all three times, it goes into leak mode, and when our crews drive by, the leak registers onscreen. In the office, that gets transferred to our leak survey crews.”

Birmingham did a pilot study with 250 of the devices, and liked the results. “The driving force for us was speed of survey,” says assistant general manager Sonny Jones. “Before the sensors, we could have leaks running for as long as two years, and that adds up to a lot of water.”

After the pilot study, BWW ordered more units and now has about 3,200 installed­. The staff estimates that 20,000 sensors will cover the entire system. Between 2004 and 2007, the sensors turned up 700 “non-show” leaks — those not visible from the surface.

Since much of the system lies in areas with acid runoff from surface coal mining, the utility must be eternally vigilant, but the sensors have helped BWW gain the upper hand.

Automated sensing is also safer. Since water lines are typically under roadways, manual leak detection is relatively risky, Goodwin observes. “With the sensors, the first pass is in a vehicle, and that reduces crew time in streets from several hours a day to a few minutes.”

District metering

The second prong in leak detection involves breaking the system down into a larger number of district metered areas (DMAs). Goodwin projects up to 200 DMAs eventually. The concept goes back at least to 90 A.D., when Sextus Julius Frontinus, water commissioner in Rome, used crude devices to measure leakage in his system.

In modern practice, a DMA is simply a discrete area of a distribution system in which flow in and out is measured automatically. Goodwin defines one advantage of DMAs: “We can measure night flows, which are less variable than daytime flows, and check them against the customer base,” he says. “If there are increases, or unusually high base rates, that’s our signal to look closer at that section of the system.”

A DMA flow rate can be compared to its own flow history, and to the flow rates of other DMAs in the system. Managers look for evidence of background leakage — the aggregation of loss from small, undetectable leaks. They also examine burst leakage — loss of water from sudden holes and fractures.

John Morrison, a widely published expert in the field, observes, “The role of DMA management is to divide the distribution network into manageable areas or sectors into which the flow can be measured to determine whether bursts are present. The duration of water being lost is kept to a minimum by analyzing the flow data so that the leakage practitioner is aware as early as possible that bursts have occurred.”

The comparisons can be used to schedule leak detection and system maintenance, and often provide good evidence to support line replacement or upgrading. The data also logically enhances a geographic information system, which BWW expects to deploy in the next two years.

Pressure management

Pressure is another critical issue. Birmingham is in the eastern United States’ Piedmont, a hilly foothill region. “We’re in 19 different gradients,” says Goodwin. “We have to pump up and down. It’s a challenge. The system has four plants drawing from three sources. Gravity and pumping combine to process 95 mgd.

“If you’ve got a hole in the pipe, higher pressure will make it leak more,” Goodwin says. “So we’re looking at flow-modulation pressure-control valves that respond to demand. As demand goes up, more pressure is allowed, but as demand goes down, in the middle of the night for example, pressure is decreased, so that you might only have 30 psi.”

Pressure management addresses maximum pressure and also surges. It has been shown that systems with intermittent supply can suffer up to 20 times the annual number of leaks compared to areas of steady pressure.

Lowering pressure as needed, while avoiding surge-related leakage, can be tricky. This is a new and promising area of nonrevenue water management for BWW. The initiative ties in logically with the DMA initiative, because DMAs help to identify where pressure management will do the most good. DMAs also serve as discrete regions where management can be applied.

“We’ve demonstrated a valve,” says Goodwin, “and we’ll be doing a pilot study.” Applied skillfully, pressure management can have a tremendous cost-benefit ratio.

A proud public agency

BWW is a proud agency. The city’s water quality has been ranked in the top five nationally. The agency also has earned the Platinum Award for Sustained Competitiveness Achievement — the Association of Metropolitan Water Agencies’ top management award — for long-term excellence in utility management.

The utility even has a world champion tapping team. Tapping teams compete to make connections to pressurized pipes in the least amount of time. The BWW team is a seven-time National Tapping Champion and won the 2005 World Water Cup Championship tapping contest in Birmingham, England.

Birmingham Water Works has been in existence since 1955. Most of the lines date to the 1920s, and a few go back to 1899. The difficult terrain, drought, the size and age of the system, and the size of the population all create challenges. Coping with these challenges with limited manpower is a tall order, but BWW is rising to the occasion by using appropriate technology and making the best use of employees’ time.



Discussion

Comments on this site are submitted by users and are not endorsed by nor do they reflect the views or opinions of COLE Publishing, Inc. Comments are moderated before being posted.