Cornell Centrifugal Pumps Solve Ragging Problems In A Washington Pump Station

Research and development at a Washington pump station solves ragging problems and reduces electrical costs by 50 percent.
Cornell Centrifugal Pumps Solve Ragging Problems In A Washington Pump Station
Five variable-speed model 8NHTA 250 hp centrifugal pumps from Cornell Pump serve the 18 mgd (design) 117th Street Pump Station in Vancouver. Pump No. 1 (foreground) ragged up in two hours even when run constantly at 100 percent speed.

Interested in Pumps?

Get Pumps articles, news and videos right in your inbox! Sign up now.

Pumps + Get Alerts

Severe ragging plagued the 18 mgd (design) 117th Street Pump Station in Vancouver, Wash., from the day it entered service in 2008.

The station, built to handle the city’s future growth, has five variable-speed model 8NHTA 250 hp centrifugal pumps from Cornell Pump. Each averages 3,500 to 3,600 gpm at 190 feet total dynamic head. Operators at the Salmon Creek Wastewater Treatment Plant in Vancouver activate one pump in the station from December to April to relieve a downstream station.

“Our situation was nobody’s fault, as we didn’t have flushable [wipes] in the early 2000s,” says Tim Scott, wastewater maintenance lead at the plant. “Ragging became a problem only in the last five to seven years.”

Even when operators ran pump No. 1 constantly at rated speed, it ragged up in two hours. When performance fell to 1,000 gpm, lag pump No. 1 turned on to lower the level of the wet well. Within two hours, it ragged up, activating lag pump No. 2. “If we ran one pump and nothing else in the control scheme, discharge dropped to 250 gpm,” says Scott.

Extended run times increased monthly power usage from 100,000 kWh for one pump to 150,000 kWh and electrical costs from $6,500 to $10,500. Scott invited Cornell Pump engineers to help isolate the problem and design a fix. “They were anxious to do research and development on an online station,” he says. Butch Veatch, municipal inside sales representative at Cornell, coordinated the eight-month project. Impeller redesign and retrofits solved the ragging problem.

Rough game

The pumps have blunt-face impellers with hexagonal lock screw heads. Impellers, the rotating component of centrifugal pumps, transfer energy from the pump motor to the fluid by accelerating it outward against the volute. Impellers have an open inlet (eye) to accept incoming fluid, usually vanes to push it radially and a bore to accept the drive shaft.

Rags sucked into the impeller eye formed a dense bundle around the lock screw head. Twice daily, two mechanics worked through a 5-inch clean-out to remove obstructions with a utility knife. “Cutting them out took an hour and cost $200 per event,” says Scott.

Because the station pumps were not cutter pumps, Cornell suggested retrofitting its standard cutter assembly — a rotating cutter ring on the impeller spinning against a stationary cutter screwed to the volute’s suction flange. The conversion included a volute machined for the stationary cutter and an impeller that accepted the cutter in place of the wear ring.

“We installed the components, agreeing to buy five cutters if the test pump ran a week without ragging,” says Scott. “It lasted two hours.” Engineers then modified the stationary cutter, thinning and profiling its three ears to influence the direction of inflow onto the cutter ring. Strike two.

To identify the flow pattern in the pump, engineers built a viewing port and mounted it to the suction flange. “We expected some weird hydraulic condition that affected how the rags hit the pump,” says Scott. However, the cloudy flow prevented viewing the cutter interface or solids entering the impeller. Large rags appeared to pass the port every three seconds and go straight into the impeller eye.

Tenacity wins out

Assuming the impeller and lock screw were the culprits, engineers reinstated the original stationary cutter and designed a smooth conical impeller with recessed hexagonal lock screw head. “When the engineers came, we’d tear down the pump, install the modified assembly and run a test,” says Scott. “Everyone was always hopefully optimistic that the next fix would work.” The experimental impeller struck out.

After two weeks of conference calls, the engineers designed a smooth cone impeller and a cone with a spiral groove. Neither met expectations. Undaunted, they designed an impeller with S-shaped vanes and recessed lock screw head. “It showed promise, but lacked sufficient geometry to guide rags into the vanes,” says Scott.

Sensing victory, the engineers extended the height and length of the vanes to create an auger that doubled as a rotating cutter. They inserted a spacer bushing to fill most of the void at the impeller eye and the deep bore for the lock screw.

“In taking an open impeller and bridging the entire face, Cornell engineers have rethought the entire impeller engineering world,” says Scott. “At the end of testing in August 2013, one pump handled all the flow. Later, it ran for a month during a high-flow event and never ragged up. The improvement is dramatic.”

In early 2014, Scott’s crew retrofitted the two lag pumps with volutes, cutter-augers and stationary cutters. “Because the impeller is also a rotating cutter, we didn’t have to buy them and saved $5,000,” says Scott. “The retrofits save $800 per day in labor and have reduced electrical costs by 50 percent. We’re seeing a quick payback on our $60,000 investment.”

Additional benefit

Weekly, the crew switched pump No. 1 to a manual cycle to clean the wet well, but the pump ragged up before it completed the pump-down sequence. After eight months of running the station constantly, solids had accumulated to a blanket depth of 300 square yards. The quote to remove it was more than $10,000.

After the pump retrofit, Scott selected three days with minimal flow to pump out the solids. The pump ran perfectly, saving the city money and enabling staff to complete the cleaning cycles.

“We couldn’t have done this without the serious support of Cornell and upper management,” says Scott. “Kay Hust, the plant manager, and our regional sewer partners — the cities of Battle Ground and Ridgefield, and the Clark Regional Wastewater District — wanted a solution. They allowed us to power the station, when it should have been offline, and solve the problem.”   



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.