Flowmeter for Low Flow in Gravity Sewer


Thread Starter


Application has 4" pipe. sewage is coming from a couple of toilet blocks in a public park. We need to accurately totalise flow, even for very low flows e.g. a single toilet is flushed.

Measuring the instantaneous flow rate is not critical. its volumetric totalising we need.

We can put a U shape section in the pipe to have section that is always full, so a mag flow meter would work. The issue with the mag flows I've looked at is the minimum accurate flow measurement is too high.

What would the experts recommend?

What a tough one - totalizing intermittent, periodic, small volume 'slugs' of water.

I was surprised when a clamp-on ultrasonic flow meter's recorded flow rate output (on a 2"/50mm line) clearly indicated intermittent, periodic blips when a night watch man flushed a toilet every 2-3 hours at an industrial building. Unfortunately, totalizing was not implemented on that particular meter installation so I have no quantitative assessment of how accurate or inaccurate a relatively low-flow flush was actually totaled in a 2" line.

I suspect that back and forth sloshing in a U section trap is inevitable, so a mag meter would need a good amount of low flow cut-off to prevent a unidirectional totalizer from totalizing forward slosh but ignoring backflow slosh. But I'm guessing that a toilet's flush flow rate (over the time of a flush) will be low enough to be very near where low flow cutoff would need to be in order to not falsely totalize sloshing in the trap.

What about dealing with the clean water line(s), rather than the sewage line?

What about counting the flushes and relying on a flow volume per flush to get total volume? With all the political emphasis on water conservation, don't the vendors spec the flow volume per flush?

A flow switch and a timer gives you raw flush counts; a shoebox PLC gives some flexibility in dealing with debounce.
Hi Morley,

I have a client who faced a similar issue, the only real difference being that he had an 8 inch sewer line instead of a 4 inch line.

Like David wisely suggested above, he installed a U-shaped trap in which to install the flanged magmeter, since it is imperative that the pipe be full in order for any magmeter to work properly.

To avoid pulse-counting problems that might arise with forward/reverse flow sloshing, he used a Neptune-brand magmeter (Tru/Mag) that provided net flow using Neptune-protocol, and used an EtherMeter (my company's product) to convert the Neptune-protocol into Modbus and Rockwell protocols for interface to his automation system.

I can probably round up some application photos, if you are interested.

I hope this is helpful!
First of all I wonder why you need to measure flow from a toilet so accurate a mag meter won't do.

Is it permissible to have a low spot in a sewer line?

Most of the flow-meters I see around our municipality are ultrasonic type from Siemens (Miltronics) installed over some sort of Parshal flume or V notch.