Many lift stations have no flow meter on their discharge, yet operators still need to know how much wastewater is arriving, both to bill for it and to study how much stormwater and groundwater is leaking into the system. An influent flow estimate solves this by using the wet well itself as a measuring tank. By combining the geometry of the well, the rate at which the level rises while the pumps are off, and the known output of the pumps while they run, a controller can work out the inflow without any flow meter at all. The technique, often called the fill-and-draw or drawdown method, turns the level signal that every station already has into a continuous estimate of incoming flow.
Influent flow estimate in one line: A pump station influent flow estimate is a calculation of how much wastewater is arriving with no flow meter installed, derived from the wet well's volume-versus-level curve, the level rise rate while pumps are off, and the known pump output while pumps are on. During pumps-off the rising level gives inflow directly; during pumps-on inflow is the pump rate adjusted by whether the level is still falling or rising. It is widely used for billing and inflow-and-infiltration studies, within accuracy limits set by the well geometry and level measurement.
The estimate rests on a simple mass balance: over any interval, the change in the volume of water stored in the well equals the water that flowed in minus the water the pumps sent out. The trick is to evaluate that balance in two phases. During a pumps-off phase nothing is leaving, so all of the rising volume is inflow, and the inflow rate is just how fast the stored volume is increasing. During a pumps-on phase water is both arriving and being pumped out, so the inflow equals the known pump discharge rate adjusted by how the stored volume is changing: if the level is still falling, inflow is less than the pump rate by the rate of fall, and if the level is rising despite the pump, inflow exceeds the pump rate by the rate of rise.
Converting a level change into a volume change requires knowing the shape of the well, and this is where the volume-versus-level curve comes in. In a simple rectangular or cylindrical wet well the cross-sectional area is constant, so a given change in level always corresponds to the same change in volume and the math is straightforward. In an irregularly shaped well, or one with benching, fillets, or a sloped floor, the area changes with depth, so the controller needs a lookup relating each level to the stored volume. Establishing that curve accurately, whether from as-built drawings or a field draw-down test, is the foundation of a trustworthy estimate.
The pumps-off phase is the cleanest source of inflow data because it isolates the inflow with nothing subtracted, and controllers often lean on it as the primary measurement, using the pumps-on phase mainly to fill the gaps. The known pump output, meanwhile, is not a fixed nameplate number: a pump's real discharge depends on the head it is working against, which changes with level, force-main condition, and pump wear. The better influent estimates therefore refine the assumed pump rate over time, sometimes by calibrating it against the draw-down slope during pumps-on phases, so that the output figure used in the balance reflects what the pump is actually delivering rather than a catalog value.
The most immediate reason to estimate influent flow is that someone needs the number and there is no meter to read. A utility receiving flow from a satellite community or a large private connection often bills based on volume, and where the connection is a lift station rather than a metered pipe, the fill-and-draw estimate becomes the basis for that billing. Because it uses equipment the station already has, it avoids the cost and the maintenance burden of a magnetic or ultrasonic flow meter on a dirty, grease-laden force main, which is a meaningful saving across a fleet of stations that would otherwise each need one.
The second major use is studying inflow and infiltration, the stormwater and groundwater that leak into a sanitary system through cracked pipes, bad joints, and illicit connections. An influent flow estimate that runs continuously reveals how a station's incoming flow responds to rainfall and to seasonal groundwater, and the difference between the dry-weather baseline and the wet-weather peak is a direct measure of how much clear water the collection system is admitting. Utilities use these station-by-station estimates to rank which sub-basins leak the most and to justify the pipe rehabilitation that reduces wet-weather load, all without instrumenting every pipe.
There is also a diurnal and diagnostic value in a continuous inflow figure. The normal daily pattern of a residential area, low overnight and peaking in the morning and evening, shows up clearly in the estimate, and a departure from that pattern, such as a steady nighttime flow that should not be there, points to infiltration or a leaking service. Because the estimate is derived rather than metered, it costs nothing extra to run once the well geometry and pump rates are known, which is why it is attractive as an always-on characterization of every station rather than a one-time survey.
An influent flow estimate is only as good as its inputs, and it is honest to be clear about where the error comes from. The volume-versus-level curve carries the geometry error: if the well's shape is not captured accurately, every volume derived from a level is off by a corresponding amount. The level measurement itself contributes noise, and because inflow is derived from the rate of change of level, small level errors are amplified into larger flow errors, especially over short intervals when the level has barely moved. The assumed pump output carries its own uncertainty, since a real pump's rate drifts with head and wear, and turbulence, foaming, and floating grease at the surface can disturb the level signal during pumping.
These limits shape how the method is best used. It is far more reliable as a running total or a daily volume, where random level noise averages out, than as an instantaneous flow reading, which can be jumpy. It performs best in a well with a clean, well-characterized geometry and a stable, calibrated pump rate, and it degrades in oddly shaped wells or where the pump output is poorly known. Recognizing that the estimate is a good volume measure and a rougher instantaneous one keeps it from being over-trusted for purposes it cannot support, while still capturing the value it genuinely offers.
This is exactly the kind of derived measurement that a cloud SCADA platform is well placed to compute and maintain. In a system such as Merobix the well's volume-versus-level curve and the pump output figure are stored as configuration, and the platform continuously watches the level and pump state to produce a live inflow estimate and daily volumes without any added field hardware. Because it holds a long history, it can smooth the estimate over sensible intervals, calibrate the pump rate against observed draw-down slopes, and flag when the numbers stop making sense, such as a well filling far faster than any storm should explain, which often means a level sensor has drifted or a pump is underperforming. The estimate then serves double duty as both a flow record and a quiet check on the health of the very sensors and pumps it depends on.
You treat the wet well as a measuring tank and apply a mass balance. While the pumps are off, all of the rising stored volume is inflow, so the inflow rate is how fast the volume is increasing, computed from the level rise rate and the well's volume-versus-level curve. While the pumps are on, inflow equals the known pump output adjusted by how the level is still changing, and the pumps-off phases usually give the cleanest reading.
It is the relationship between the water level in the wet well and the volume of water stored at that level, which lets a level change be converted into a volume change. In a plain rectangular or cylindrical well the area is constant so the curve is a straight line, but in a benched or sloped well the area varies with depth and a lookup is needed. The estimate is only as accurate as this curve, so capturing the true well geometry is foundational.
It is more reliable as a daily or accumulated volume than as an instantaneous flow, because level noise averages out over time but is amplified when computing a rate over a short interval. Its accuracy depends on how well the well geometry is characterized, how stable and well-calibrated the pump output figure is, and how clean the level signal stays through turbulence and grease. Used as a volume measure in a well-characterized station it is good enough for billing and inflow-and-infiltration studies.
Merobix reads your field devices into a cloud SCADA - the real thing behind these terms, live in days from any browser.