Thousands of water and wastewater treatment facilities use granular media filters. An operational control system – along with media configuration, underdrain system, and backwash process – is an important consideration because it determines how water flow is controlled through the filter.
Four basic types of operational control systems are used in gravity filtration, with some variances from plant to plant. Each has advantages and disadvantages.
- Effluent rate of flow
- Influent flow splitting
- Constant level
- Declining rate
Selecting the best filter operational control system option entails understanding which one works best in unison with the overall plant design. Filter control considerations should be taken into account with each redesign, or in the initial stages of development if a new plant is being constructed.
For existing facilities, understanding how operational control systems work is important because it helps operators improve water quality and maintenance, as malfunctions can damage a filter bed and allow sudden changes in the filtration rate.
1. Effluent Rate of Flow Control
This type of system controls the rate of treatment through a cell on the filter’s effluent or discharge. Basic components include a flow-sensing device, rate controller, and modulating valve. A wide variety of flow-sensing devices may be used, including direct reading meter systems and indirect reading systems such as Venturi or orifice plates.
The rate control design also varies considerably, depending on the flow-sensing device and modulating valve design. Signals from the flow sensor, rate controller, and valve may be pneumatic, electric, mechanical, or a combination of these.
The flow-sensing device is used during operation to determine the flow rate through the filter effluent line. The measured rate is directed to the rate controller, which compares the measured flow rate to an operator-adjustable set point. If the measured and set-point values differ, the controller sends an appropriate signal to the modulating valve to open or close, thereby aligning the measured and set-point flow rates.
When a filter run starts directly after backwash, the actual head loss across the filter is much lower than the available head loss. Available head loss can be designated into two simplified components: clean-bed head loss and dirt head loss. Clean-bed head loss consists of the filter system’s fixed portions and includes hydraulic losses through the media, underdrain, and piping system. Dirt head loss is the pressure loss across the filter associated with captured solids in the filter media.
Following a backwash, when dirt head loss is at its lowest, the filter system has the least amount of restriction; if left unadjusted, water will flow at its fastest rate. To control the actual rate to the set-point value, the modulating valve is closed to induce head loss through the valve to restrict the flow back to the set-point rate.
As dirt head loss increases, the flow modulating valve is gradually opened to maintain a constant flow rate leaving the filter. The level in the filter cell may vary over time, and a backwash can be initiated when the actual head loss reaches the available head loss across the filter – if time or filtered turbidity doesn’t dictate the frequency. The point at which actual and available head loss equal one another is indicated when the cell water level reaches the maximum elevation and the modulating valve has moved to its full open position.
- The allowed flow rate through a cell is held constant at the set point at all times despite fluctuations in head loss through the system.
- Effluent rate control systems allow an operator to easily control the maximum flow rate through a single filter cell.
- The constant treatment rate also prevents dislodgement of captured solids within the filter bed, which can cause breakthrough to occur.
- Effluent rate control systems require the most operator attention to calibrate instruments involved. Overall plant system control is especially important with such systems to prevent multiple control schemes from fighting against one another.
- This type of control lends itself to a start-and-stop operational scheme because the flow to multiple filters isn’t regulated. This can develop a slug flow condition in which the clean filter is slugged with a large initial flow of dirty water. To account for this type of condition, a filter-to-waste step is often incorporated in the filter. When a filter is to be put into backwash flow, the unit is stopped by the effluent valve – not at the influent. If the system fails, the filter will operate as a declining-rate filter.
2. Influent Flow Splitting
An influent flow splitting filter system has an adjustable weir positioned at the entrance of each filter cell. The influent weir is the system’s main component. A feed pipe or flume common to all filter cells carries the water to the individual weirs. The inlet weirs are adjustable so each can be positioned at the same elevation to obtain uniform flow splitting. Ancillary equipment often includes an influent control valve and an effluent hydraulic control point to maintain a minimum water level in a filter cell. The influent control valve is used to stop incoming flow if the cell needs to be removed from service, such as during a backwash event. The effluent hydraulic control point may consist of a downstream weir or upturned loop in the effluent pipe.
As water passes over the influent splitting weir at each cell, an equal amount of water is delivered to each online filter cell. The effluent hydraulic control point is included to prevent surface scouring or dewatering of the filter media. The hydraulic control point is often set a few inches above the resting filter media surface. The flow from the influent weir may be directed through the wash-water collection troughs to minimize media scouring directly after backwash when the dirt head loss and filter cell water level are lowest.
As dirt head loss increases, the cell water level will rise within the cell. Terminal head loss is located at a water elevation below the influent weir to prevent the weir from flooding, which would make it ineffective for flow splitting.
- If a cell is removed from service, the influent flow to the filter cells is automatically divided among the remaining cells in operation, with the water level increasing in depth over the weirs. For multiple cell operation, the backwash sequence can be adjusted such that the need to backwash one cell can be offset from the remaining cells to prevent wash-water recovery systems from being overloaded.
- No mechanical or electrical components need to be calibrated or adjusted to maintain system functionality.
- Typically the weirs are located so plant operators can visually inspect them to ensure debris hasn’t fouled the weirs. Visually observing the water level in the cells also allows operators to easily see when a cell is nearing the need for backwash.
- It’s possible to exceed the filter cell’s designed hydraulic loading rate if incoming flow exceeds the available online filter area. This situation may occur through an increase in flow to the filters, multiple filter cells are offline at one time, or a combination of these. If the influent flow varies, the hydraulic loading rate in each cell will also change.
- Sudden changes in hydraulic loading rate can dislodge captured solids, causing turbidity breakthrough.
- The design of the downstream hydraulic control point is also important to prevent scouring of the media surface, causing differences in media bed depth that can result in uneven filtration rates through the bed.
3. Constant-Level Control
This type of filter control uses a level-sensing device in each filter cell to communicate with a control valve in the effluent line. The communication method can take various forms, including mechanical linkage, pneumatic pressure, electrical signal, or a combination of these. A variety of level-controlling devices may be used. The most common is a butterfly valve equipped with an appropriate actuator to position the valve disc.
Valve size selection is critical for operation, and selection should be made such that a small change in disc position doesn’t drastically change the flow through the valve.
Constant-level control is similar to effluent rate of flow control, as the sensing system compares current conditions with desired set points and works in conjunction with the control valve to align the two conditions. In a constant-level control system, the level-sensing device determines actual water level and compares it with the set point. The comparison method depends on the sensing device, control valve actuator, and communicating controller.
Like the effluent rate control system at the start of a filter run, when dirt head loss is low, the control valve opening is reduced to impart head loss to the filter system. As dirt head loss increases, the control valve is opened to balance the overall filter head loss. Terminal head loss is achieved when the effluent valve reaches the full open position and the dirt head loss continues to increase, causing the filter water level to increase and indicating the need for a backwash.
Another common option is to use influent flow control, which is similar to an effluent rate control system except the components are located on the influent side of the filter.
- Eliminating potential media surface scouring is a primary advantage of constant-level control systems, because the cell water level is always located well above the media surface.
- Like effluent rate-of-flow control systems, constant-level control systems also allow the use of negative head loss, or vacuum suction, across the filter system to extend filter run time.
- Provided the influent flow is controlled at a constant rate, a constant-level control system will maintain a constant rate of filtration.
- Switching back and forth between flow control and rate control can happen at hydraulic breakpoints.
- Constant-level control systems can have potential flow stabilization issues.
- The operator is not able to visually see the head loss building in the filter.
4. Declining-Rate Filtration
Declining-rate filtration is one of the oldest and simplest methods of filter control. It is used with multiple filter cells – the greater number of cells, the better the performance. The system’s main component is the filter media.
To limit the maximum rate of filtration through a cell, a flow restriction may be placed in the effluent pipe or, in some special cases, on the influent side. The flow restriction may simply be an orifice plate.
A two-position control valve may also be used where the valve is partially closed directly after a backwash to restrict flow of the cleanest cell.
The easiest way to understand declining rate filtration is to consider the entire filter system rather than a single filter cell. The influent water enters the cell below the cell water level, allowing the water to flow freely into the cells. All filter cells are fluidly connected through the influent system. With the filter influent unrestricted, the water level will be the same in each filter cell. After the initial system startup, the filter cell backwash events are staggered such that each filter cell is in a different state of cleanliness.
Therefore, flow through a cell is based on the dirt head loss in a filter cell, because all cells have the same water level above the media. Thus, the cleanest (most recently backwashed) filter cell will pass more water than the dirtiest cell (closest to backwash).
As dirt head loss increases across the system, the level in the filter cells will increase until the maximum system head loss is reached. Because all cells have the same water level, the filter cell that has been operating the longest is then backwashed.
After the backwash event is completed, the system’s overall dirt head loss is reduced, so the cell water levels will drop and the cycle is repeated.
A flow-restriction device, such as an orifice plate or partially opened valve, is often installed in the effluent line to limit the maximum rate of filtration through a cell. This flow restriction is used to limit the initial flow passing through the filter – essentially a speed limit for the filter.
- Because of their minimal control component requirements, declining-rate systems are fairly easy to operate.
- Declining-rate filtration allows better use of the filter media depth. When the media is cleanest, the higher filtration rate drives solids deeper into the bed. As dirt head loss increases, the filtration rate decreases to minimize shearing forces within the bed.
- The system handles variable influent flow rates better than most other control systems, as the variations are spread out among more filter cells through the common influent system.
- There is the risk of turbidity breakthrough in clean cells if not restricted or limited to a maximum rate. This can be a regulatory issue, as most regulatory agencies require the use of a maximum filtration rate.
- Individual filter cell head loss development can’t be visually observed when multiple filter cells are present. The influent flow, common to all of the cells, hides the fact that one unit may have more head loss than another. Fortunately, today electrical monitoring systems are readily programmed to track cell run length to properly sequence backwash events.
NOTE: The descriptions provided here are based on a single filter cell in operation and don’t account for upstream or downstream processes and their control systems. In operating a filtration system, it’s important to consider how a filter’s operational control system affects or is affected by control processes throughout the plant.
A longer version of this article originally appeared in Opflow in November 2016. Source: Pallwitz, S. & St. Germain, D., 2016. Granular Media Filtration: Maintaining Operational Control Systems. Opflow, 42:11:10. Adapted and republished with permission from American Water Works Association.