Stay Proactive with Sanitary Sewer Flow Monitoring
In a perfect world, stormwater would never find its way into a community’s sanitary sewer system, let alone, to a treatment plant. However, inflow and infiltration issues are real-world problems that require a complex monitoring process. Implementing flow monitoring inspections identifies problem areas of inflow and infiltration in a sanitary system and helps determine the proper steps for establishing a repair plan.
The process of collecting adequate flow data is a challenging undertaking that if performed incorrectly, could result in costly mistakes. In this podcast, Brett Paige, PE, takes an in-depth look at the flow monitoring process, the techniques available, and some of the ways our experts ensure you’re getting quality data to achieve the best results.
- What’s Included in the Flow Monitoring Process? (0:19)
- Potential Flow Analysis Techniques (2:22)
- Inflow Coefficient Method (2:41)
- RTK (4:10)
- Capacity Evaluation (5:30)
- Mannings Equation (5:52)
What’s Included in the Flow Monitoring Process?
What’s included in the flow monitoring process are the flow media equipment and rainfall gauges. A big point that I want to stress is if you’re going through a flow monitoring project, don’t just rely on local and national reported data such as airports. It makes it difficult for us to process that information and understand it. What we’re trying to do is look at a system response, and if we’re getting rainfall-reported data that is 10, 15 miles away, anything that’s not local makes it very difficult. I think in Iowa specifically, we see a lot of very isolated thunderstorms. I’ll give an example here in Des Moines. We can get a thunderstorm that maybe hits the Ankeny or Urbandale area and gets two inches of rain. I live down on the south side, and maybe it doesn’t rain at all. That just kind of visualizes why it’s important to have a local rainfall gauge. Typically, we want a rainfall gauge for the flow monitoring project that’s within five to 10 square miles per gauge.
Delineating subsystems to target smaller sewer sheds, just trying to pick out where our meter should be to isolate pieces of the system off as we look to prioritize areas that might have higher I/I than others and where we want to start with our inspections. These are just some general recommendations here for some approaches that we typically look at at the high level when we’re looking at planning, a flow monitoring project. One meter to every 40,000 feet of sewer, there are a lot of variables that go behind the scenes that depend on urban density. There are a lot of commercial, industrial, and residential properties. There are certain values that we try to target, but really what we’re trying to do is establish enough flow that goes across our flow meter that is laminar, non-turbulent, that’s not restrictive from the surcharge, and that’s given us enough velocity. So we get a decent amount of accurate data as we collect the flow monitoring component. And then the monitoring period. I think just understanding the goal here. If we’re trying to establish how much clear water is in our system, we want as wet weather activity as possible. Typically, we’ll do a three-month monitoring period, whether in the spring or the fall, depending on where you’re located, but trying to capture some thunderstorm data to project out some of that peak inflow and those infiltration values.
Potential Flow Analysis Techniques
I’ve got a couple of flow analysis techniques that I want to go through at a high-level overview of what’s out there and what we typically look at. The main goal we’re trying to quantify sewer flows, dry weather, inflow, and infiltration. We’re using that rainfall, depth, and velocity to create ratios and project out values for different theoretical storm occurrences.
Inflow Coefficient Method
We’re using graphical displacement information to understand thousands and thousands of rows of data and just creating that graphical output that we can understand from the engineering level, what the flow data means, and then existing in future projections. Inflow coefficient Q versus I, we’re referencing the rational formula for runoff, and we’re replacing that runoff coefficient with the KP value. That KP value is representative of the average coefficient solved, based on the storm characteristics. So, you’re getting a ratio of the flow, and the intensity in the area, and we’re selecting storms that have a certain KP value. Other criteria that we’re using to select storms are going to be based on the amount of depth of rainfall that we received over the sewer shed. Typically, we’ll look for anything greater than a quarter-inch just because we want to see that system response, and you can kind of get some diluted data if you include anything less than that. It does affect your trend projection as we plot Q versus I.
For those that aren’t familiar with those theoretical storm events, typically, we’re looking at projecting out up to a hundred-year storm event, which would be rainfall that here locally in Des Moines would be over seven inches of rain in a 24-hour period. There’s a 1% chance of that occurring in any given year, so we’re trying to optimize what we need to size our system to. I think sizing to a hundred years is economically difficult. So, we’re typically targeting a lower frequency storm to do that and this is just allowing us to understand the capacity of our system with these projections.
RTK is another method that we will use to evaluate peak flow in our sewer systems. This is an accepted EPA-approved method. They have a toolbox that you can access online that has predeveloped spreadsheets where you can input data to help with some of the analysis.
What do these acronyms mean? The R-value is going to be the fracture of rainfall over the sewershed that enters the collection system. T is the time to peak, and K is going to be the ratio of time to the recession, to time to peak.
How this system works, it’s essentially a spreadsheet tool that we can go in and manipulate these values within reason. There’s typically a starting point that we like to see these values within the range. So we’re going in and manipulating these fractions and ratios to try to get something that matches up to the overall observed hydrograph. This is a widely accepted process. It’s great to help with our modeling inputs and just about with anything, you know, the more experience you have gone through this process, the quicker it can be.
So what are we trying to do with this data? We’re trying to quantify base flows for our dry weather periods, establish our peaking factors, how much I/I is in our system, and project-out theoretical rainfall events to understand capacity limitations. In future planning, if we want to expand to handle a larger frequency event, maybe understand some of the I/I reduction that we get from pre and post-flow monitoring from rehabilitation. Determine our criteria for model inputs and evaluate limited capacity locations.
Jumping into capacity evaluation, and some of the basics. What are we trying to design for? Average daily flow to handle the low flow periods daily flow plus infiltration. So kind of a normal spring month, maybe that groundwater’s elevated, and we’re handling that small fraction of excess clearwater with our wastewater production. Then peak flows look at RDII plus all the above. So how does our system respond under all those scenarios?
Manning’s equation. Not going to go into a whole lot of depth with this, but if we’re looking at a singular pipe network, this is what we’re designing off of before we get the pressure flow, and these are the variables that we see. This is good for a small handful of pipes, and a point to make varying slopes are going to result in varying capacity. If we steepen up a pipe it’s going to result in more flow capacity of that piping system. And one of the things that make this difficult to employ on a larger scale, if we’re doing a system-wide capacity evaluation, the iterations that it would take to do all that by hand are just almost unimaginable from an effort standpoint.