One advantage to the COVID-19 national shutdown – it gave Environmental Works, Inc. (EWI) a chance to work on maintenance projects, including an upgrade to Willard Central Elementary’s aging wastewater treatment system. While students, teachers and administrators were away in spring and summer 2020, Jon McKinney, associate engineer in EWI’s Investigation/Remediation Department, constructed a new control panel with a programmable logic controller (PLC) for the four-pump wastewater treatment system. A PLC is a digital computer which has been ruggedized and adapted for the control of industrial processes.
EWI’s Field Services Department has been servicing Willard Central’s wastewater treatment system for over 20 years. During that time, the system control panel, which employed a basic timer running to a pump duplexer, had gotten worn down and became inoperable. The duplexer switched between the two sets of pumps, running two of the four pumps for a set time, to control loading of the treatment system.
All wastewater from the elementary school flows into the treatment system, according to McKinney. A screen keeps the solids in the septic tank, which are pumped out periodically, while the liquids move on and are pumped through a recirculating media bed. Every time the pumps switch on, they push the wastewater through lateral pipes buried six inches from the surface of a three-foot-deep pea gravel filter bed. Running along the lateral pipes are tiny holes with loose fitting caps covering them. The wastewater sprays upwards through these tiny holes, hits the caps and disperses downward into the filter bed.
The gravel filter bed is bioactive, according to McKinney. “Atmospheric oxygen permeates into the filter bed so the bacteria growing on the rocks can use the oxygen to eat the waste,” he said. “Rather than it being a constant filtration process like drinking water filters, it’s a bioactive bed that you have to periodically load (with wastewater) and let rest.” This provides the necessary environment for treatment in the gravel filter bed.
The filter bed features sixteen lateral pipes in total, with eight pipes loaded with the initial pump cycle, and the remaining eight loaded during the alternating cycle. “So, the loading for each zone of the filter bed is actually half of the pump run time,” said McKinney.
The wastewater filters through the gravel bed to an under drain, which carries the water to a mechanical valve that controls how much water goes back into the re-circulation tank to be loaded into the bed again and how much gets chlorinated for disinfection and then discharged. “It’s random, but on average the water might be reloaded fifteen times before it gets discharged,” said McKinney.
The operation of the original pump timing control system initially made sense to McKinney. “At first I was like, that’s really a simple solution,” he said. Then, like a true engineer, he envisioned different scenarios where that basic solution didn’t provide for optimal control and treatment of the wastewater.
“One of the biggest challenges to operating this system in the past is the amount of variation it sees,” he said. “It’s not like a truck stop or a subdivision that has consistent usage. It’s much more variant – normal load, summer break, winter break, school events with parents present, etc.”
The new PLC controlling this system has eight inputs and controls four outputs – one for each pump. The pump chamber has three probes that correspond to different tank levels from low to high. The probes are the inputs to the PLC and tell the system when the water levels are too low, when the water has shifted from a low load to an average load, and when the water has increased to a high load.
“Say you get to the low level and the low level shut-off stops the pumps. So, you’ve got this sensor and the water level drops below it. The pumps shut off and a little bit of flow is still coming into the system, then the water level goes up to the sensor indicating a level sufficient to turn the pumps on. When the water drops below the probe, it turns the pumps off, a little water comes in the tanks, it turns the pumps on, draws it down past the probe,” McKinney said. “There is the potential for the signal to just sit there and bounce like that. So, I configured it so that if a low-level gets triggered, there is no pumping for 12 hours, which gives some time for some water to reenter the tank, unless the middle probe gets hit, which overrides that 12 hour delay.
McKinney provides an example of when a middle probe might get triggered during a 12-hour no-pump event: “Say it’s registration day in the summer. You’re in the low mode due to summer break, but then at 8 a.m. on registration day, parents and students start showing up and loading the tank. You would want an override so that if the water level in the tank rose to the middle probe, it would go into that pumping mode rather than waiting 12 hours even though the tank’s filling up. I did the same thing for the high-level probe.”
Event counters tied to each of the probes were programmed to track and record tank level changes and to help fine-tune the run timers for the pumps to keep the system operating in the optimal range for the observed loading. For the first month and a half after installing the PLC, McKinney visited Willard Central weekly to check on the system. “And we slowly adjusted the timers according to the event level logs,” he said.
With this knowledge, McKinney programmed the timers accordingly: “Anything between the low and middle probes, we have the timers to pump for four minutes and off for 56 minutes. That’s one pump out of each tank and it alternates each cycle. One pump from one side and one pump from the other side on for four minutes and off for 56 minutes. If the water is between the middle and high probe, two pumps are on for four minutes and off for 26 minutes. If it gets to the high probe, all four pumps run for four minutes and are off for 26 minutes. And it does that for at least two hours, unless the water level drops down to the middle probes again.”
EWI recently took the first effluent sample from the wastewater treatment system after the new control system was fine-tuned and had complete removal of biochemical oxygen demand, according to McKinney.
Biochemical oxygen demand (BOD) is a measure of the amount of dissolved oxygen in water used by microorganisms to decompose the organic matter present in the water. This parameter is used as a general measure of the degree of organic pollution.
“We had non-detect BOD coming out, so full removal of organic waste with the pump timing optimized,” McKinney said. “The controls are working right. The loading is working right. It is distributing water evenly into the bed. Full removal of BOD indicates the system is operating very efficiently.”
McKinney said he will continue to adjust the timers if needed. “We can view a log of all the level events happening with the PLC and adjust the timers accordingly,” he said. “Now, what I’d like to see, is during low load periods, winter break and summer, that the water remains between the probes.”
“If there’s a large event at the school, we might get a high level where all four pumps are running. If that only happens a few times year, that’s fine. And when we’re on a light load, like on summer or winter break, it’d be great with the run time adjustments if we could get the water to stay between the low and middle probes. Then we’ve got the tanks always at a reasonable level and the PLC deciding which mode to run to maintain optimal dosing for a given load.”