As part of the continuous monitoring of our network, we have equipped a measuring point with an ultrasound level/Doppler velocity sensor.
The problem we have is that this section in particular is susceptible to sedimentation. When that happens, the velocity sensor doesn't work and, in any event, the hydraulic law becomes skewed.
The main cause is the siphon just upstream from this point + an oversized interceptor.
Installing the sensor upstream of the siphon is impossible because the manholes are located in the middle of a roadway or in a roundabout/traffic circle. Installation in the manhole downstream is equally impractical as another branch of wastewater feeds into the interceptor.
I await your input!
Having taken part in setting up the first network with direct monitoring—real-time and continuous—of interceptors in the early 80s, it soon became clear that measuring the flow rate was arbitrary precisely because the problem of sedimentation itself occurs randomly.
Assessing water height in the interceptors is no easy task but is usually enough to determine whether the sewer is surcharging and creating an overflow, which seems to me to be the point of continuous monitoring.
Managing a wastewater network is less about monitoring flow than water height, because an overflow can occur at flow rate X one day and at X/3 the next if there is fouling in the interceptor.
The goal of continuous diagnostics is to check whether there is overflow in the collection pipes and when. If you know the spot height of a stormwater overflow, using the water level (reference altitude), you can determine if it is too high and therefore whether or not there is an overflow. In some cases, you can even calculate an approximate value for the amount of water discharged.
I would suggest:
- moving your measuring point to a more relevant section of the network that is also less susceptible to fouling and other deposits.
- settling for the height of water based on elevation (spot height)
- or installing a grit-removal chamber upstream from the measuring point, which is a bit complicated and old-fashioned but a really useful way of ensuring your operation runs smoothly.
Thank you for your reply. Indeed, there is an element that I neglected to mention. We operate a separate sewer network. So, the objective is to measure volumes passing through this particular branch of the network (there is no overflow or weir).
I had not thought about a grit removal chamber, but that might be the solution to our problem.
Changing the measurement point does not seem viable (see previous explanations).
If you think of any other cheap and cheerful solutions, I’m all ears.
Your objective seems to be almost impossible to attain, especially if you require accurate measurements.
Grit-removal chambers are reserved for combined or storm sewers. If you measure a wastewater interceptor, there should be no sand, so no need for a grit chamber.
Another solution that would give an approximate value would be:
(1) Cleaning the network
(2) Measuring flow for a week using a temporary portable device to calibrate the ratio of water height to flow rate.
(3) Then noting only water height and deducing an approximate flow.
A simple software program could later determine any flow drift (water level), by comparison with the calibration curve, thereby triggering a “cleaning” alert for any sediment encountered at the measuring point.
This would require frequent cleaning interventions.
I still don't see the point of measuring flow rates on a separate sewer: the network user meters give a good idea of the volumes flowing through the pipes. And mobile devices are generally sufficient to identify flow patterns (you can even use "fuzzy logic" applications)
But if the interceptor is accumulating grit and fouling, then it is separate in name only is it not?
Of course, if we are encountering sediments, then that means the network is not strictly separate, but it is supposed to be! What's more, road construction is underway on this stretch of the network. That accounts for some of it.
I am currently monitoring sedimentation on a daily basis to try to determine a new hydraulic law.
I agree that we will have to perform more regular cleaning of this section.
In response to "the point about measuring flow rates on a separate sewer," it is simply a regulatory requirement for treatment plants of more than 10,000 PE. Monitoring also allows us to see the effect of parasitic clear water (seepage or meteoric) and then to perform the necessary interventions based on the results—all this in order to improve the quality of the wastewater system. The number of connections indeed gives us a gross estimate as to the load in this run of pipe, but I think we can all agree that theory and practice don’t always match.
As you are operating a separate sewer, my advice would be to wait until roadwork is complete to see if the grit problem persists.
But you tell me that runoff is not the only issue: there might be parasitic water inflow. As in many places I think—so, the network is not strictly separate.
The presence of parasitic inflow may be suspected based on two parameters: in general the relationship between the night min and daily max flow rates (dry weather) must be less than 10 on a strictly domestic network. Or, take a look at the daily and nightly COD and get an approximate dilution coefficient and therefore a dilution rate.
I still do not understand why your local water authority requires flow rate measurements for a separate sewer whose quantitative variations in wastewater (except for daytime/nighttime variations that would have been taken into accounting during the planning phase) are related to the population, seeing as how the number of users is likely to change slowly in one direction or the other.
Obviously, if this is a quasi combined-sewer with a flow rate that is directly related to rain, that might explain things.
A strictly separate network does stay that way for long: it needs to be checked every ten years (smoke; field measurements).
Speaking from experience, reliably measuring flow rates other than via calibrated devices (and even then!) is risky, and the degree of accuracy (value given by the measuring device differs from real value by less than 20%) does not exceed 5% at best.
However real-time management of the height of the flow at different points on a network provides a wealth of information on its condition and any anomalies that may arise.
If the problem is regular fouling as a result of a low slope or a decrease in flow speed, you might try flushing. It is a bit old school, but it may fix the problem if it is a local one. This would enable you to regularly flush any fines that might be deposited at a right angle to your sensor. NB: You will likely have to increase the frequency of cleaning downstream so as to avoid simply displacing the problem, but I think it will improve your readings.
Hoping to have contributed to the discussion.
Thank you for your input.
The interceptor is D600, so I don't know if that method applies. From what I read, flushing is effective in pipes below 300/400 in diameter.
What do you think? Do you have feedback on this diameter for an interceptor?
I don't know the configuration of the network/manhole nor the expected flow rates.
Would it be possible to create a channel to measure water height at the bottom of the manhole (simply by narrowing the section that runs its length) in order to force a difference in elevation between the water upstream and downstream of the manhole?
This might even boost flow enough to dislodge some of the fouling?!
Indeed, we have looked into that.
Increased speed for a small section would enable us to flush out the grit.
It could be done simply by installing a Venturi system. In theory, the sensor would be only slightly disrupted without cutting out downstream. It remains to be seen how the current would react in real life.