Without spending money or hiring an outside contractor, the City of Chandler, Ariz., increased production of its drinking water treatment plant by 4.7 percent while decreasing annual electricity consumption by 1.4 million kWh, reducing well usage, reducing disinfection chemical usage, and saving more than $1.1 million. The key to the project's success was to involve the water treatment operators throughout planning and execution.

Anticipating the U.S. EPA's 2012 disinfectant byproduct rule, Chandler's Water Quality division and Victoria Sharp, water systems operations superintendent, began looking at water age two years ago. Some reservoirs in Chandler weren't experiencing good turnover, and that was affecting water quality.

Water has naturally occurring total organic carbons (TOC), Sharp observes, and when a disinfectant contacts TOC, it produces trihalomethanes (THMs). The longer the disinfected water stays in a reservoir, the higher the potential for THM formation.

Sharp wanted to eliminate THMs by getting better water turnover in the reservoirs, but improving the system would mean changing the team's philosophy on water treatment. "Our mentality was to keep the tanks full at all times using well water," Sharp says. "That's how we had always done business. But we needed to change so we were using the water we had instead of just filling the reservoirs."

Growing demand

A suburb of Phoenix, Chandler covers 77 square miles and has a population of 240,000, up from just 30,000 in 1980, at which point large companies began moving into the area, making it one of the fastest growing cities in the nation. To keep up with the growth, Chandler added wells and waterlines. The city now has 27 active wells.

Chandler's traditional water treatment plant, which was upgraded in 2008 to handle higher capacities, can process 60 mgd. A treatment plant shared with the neighboring town of Gilbert provides an additional 12 mgd as needed. The city's average demand is 54.5 mgd.

With so many field sites scattered around the city, each location had a different water turnaround rate based on demand, water pressure and elevation. In addition to concerns with water age, Sharp saw an opportunity to use more surface water instead of pumping from the wells, thus saving significant on water resource costs, electricity, and disinfection chemicals.

Team buy-in

Instead of hiring a consultant, Sharp went directly to the 10 operators and maintenance staff members for advice on how to optimize system efficiency. Asked for buy-in at the beginning, the team members owned the project and were inspired to make it work.

"They started talking among themselves and pretty soon it grew," Sharp says. "They were coming up with ideas on what to try. Suddenly there was a new enthusiasm in the group that hadn't been there in years. We were trying something different."

A team of 40 people in the Water Resources, Water Quality and Engineering departments contributed to the project. They experimented with novel ideas, executed test runs and continuously monitored the system.

For example, operators had a theory that the best way to get a hydraulic grade line (HGL) for each field site would be to turn off all the pumps in the system simultaneously during two consecutive low-demand evenings, while operating the large Simflo finished water pumps at the surface water plant at the known HGL pressure point. The plan proved worthwhile, and the team accurately identified the correct HGL pressure set points at each site.

Automating the process

After 12 months of experimentation, the team enacted a new process for operating the drinking water treatment system in January 2011. Previously, operators manually turned on pumps at each site via a Foxboro SCADA system at the plant. Pumps often ran all day.

"Pumps would run constantly in the old days," Sharp says. "Every day when people were waking up and there was demand, operators would turn on pumps as needed, then turn them off later in the day when demand was generally lower. It was a system that worked but wasn't very efficient. They were operating blind for a lot of years. They didn't know what the pumps or equipment were doing."

Now the system automatically monitors the pressure setpoints at each site and adjusts accordingly. When the city pressure gets low due to high demand, the reservoirs throughout the city are automatically pumped, draining the tanks that were filled during the previous night shift. When city water demand drops off and the water pressure rises, the booster stations shut off.

The operators then raise the pressure setpoint at the surface water treatment plant slightly and begin the long process of refilling the off-site reservoirs with surface water from the water plant before the next hit. This process of filling reservoirs is necessary because there are no water transmission lines to the reservoirs in the city's largest pressure zone requiring all the reservoirs to be filled using the distribution piping system.

With pumps running only when needed, the entire system runs more efficiently. Operators closely monitor the performance of each pump for an optimum output of 97 to 100 percent.

"We have saved significant electricity by not running pumps that aren't necessary," Sharp says. "We are only using the pumps necessary to push out as much water as the system is calling for. If a pump drops below 90 percent, it generates an alarm, and the operator can decide whether to turn off that pump."

An optimization chart helps operators determine if a pump is running efficiently and whether it should be turned on or off. Having an overview of the entire field provides more information to support better decisions. Operators work strategically as a team to fill up multiple tanks simultaneously without negatively affecting pressure in the delivery system. "There's more strategy now than there was before when we were just turning pumps on and off and flying by the seat of our pants," Sharp says.

Financial sense

By filling the reservoirs only as necessary, Chandler can use more surface water sources instead of pumping from the wells. The cost of buying surface water from the Salt River Project is $44 per million gallons, while running wells costs $394 per million gallons. Some wells that required arsenic treatment cost $759 per million gallons.

Using pumps more efficiently and less often saves Chandler $146,000 a year. The reduced power usage eliminates emission of 996 metric tons of carbon dioxide. In addition, the city has saved $931,000 in annual water resource fees by switching to surface water, while still producing 4.7 percent more water than in years past.

Since well operating conditions have been optimized, the city has saved $55,000 annually in calcium hypochlorite tablets at the well sites. In total, the city has saved $1,132,000 in energy, chemicals and fees without investing in new technology, all thanks to the operators' quick and smart decisions.

"Once it was the operators' project, everyone was into it. It went like a house of fire," Sharp says. "Everyone wanted to get into it. It was amazing. We had no idea that we would be able to accomplish as much as we did in such a short amount of time."