Detection of leaks: next-generation pipe inspection

Abstract

The wastewater system in the Netherlands is ageing, which increases the probability of failure of the systems as time progresses. A vital failure mechanism is leakages, which occur in leaky connections, joints, or at cracks and holes along the pipe. Eventually, leakages result in infiltration of groundwater into the sewer pipe or exfiltration of sewage into the groundwater system, depending on the location, condition, and system type of the sewer. When the sewer pipe is located below the groundwater table, infiltration of groundwater will occur, and when the sewer pipe is located above the groundwater table, exfiltration of wastewater will most likely occur. Both can have severe consequences for the sewer system, e.g. collapsing pipes, reduction of the groundwater quality, silting up the sewer, and high hydraulic influent load. To prevent these failures, maintenance of the urban drainage system is essential, also referred to as ‘asset management’. It is crucial to find out whether there is a leak in the sewer pipe, where the infiltration or exfiltration occurs, and what the amount of leakage is.

There are different methods to detect leaks, such as camera inspection, pressure tests, and distributed temperature sensing. However, these different techniques each have their specific limitations. The Focussed Electro Leak Location (FELL) system, which is the main focus of this report, has been developed to detect and locate leakages by using electric voltage between an electrode in a sewer pipe and an electrode in the surrounding ground. The electrode in the sewer pipe, also called the probe, consists of three parts. The outer electrodes serve as a guard and also to send electric voltages, like the middle electrode, but they are mainly there to block the area behind it so the signal does not escape. This probe is pulled through the sewer. As soon as an increase in current is visible, this means that the resistance in the sewer pipe wall decreases, and there is a risk of leakage. There is some noise in the signal, but generally, when there is a leak, a more prominent peak in current is present. This peak can say something about the length of the leak, it could say something about the size of the leak, and perhaps it could say something about the leakage rate. There are doubts about whether the last is possible.

Explorative experiments with a PVC pipe above the surface, under ideal conditions, have shown that a clear signal can be obtained caused by leaking water using the FELL prototype 2.0 equipment at hand. The results show that the settings of the FELL prototype 2.0 have a major impact on signal size and noise. However, based on the performed field experiments with a concrete pipe below the surface, it can be concluded that also passing trains, moisture of the soil surrounding the grounding pin, and the position of the FELL prototype 2.0 in the sewer pipe have a strong influence on the resulting leak detection signal of the FELL prototype 2.0. The signal-to-noise ratio was ~1. Environmental factors influence the results in such a way, to an order of magnitude of even greater than 0.1 mA, that it is impossible to see where and whether there is a leakage in the concrete sewer pipe below the surface. To make the signal visible in practice, the FELL prototype 2.0 has to be redesigned, so that environmental influences have limited and known influences on the overall measurements. Also, the vertical movement of the FELL prototype 2.0 during the pulling of the system through the concrete sewer pipe has to be avoided as much as possible, as any change in the measuring geometry clearly affects the obtained signal. This research showed that the FELL prototype 2.0 should best be positioned precisely in the centre of the pipe. In short, the measurement principle theoretically sounds promising under ideal conditions, but the technical implementation of the FELL prototype 2.0 is very fragile, as shown in this study. Theoretically, it is possible to detect leakages, but in a practical sense, many improvements still need to be made to make the implementation robust. It can be confirmed that the performance of the FELL prototype 2.0 was not sufficient yet to tell if this device is able to detect, locate and quantify leaks under simulated field conditions. Both the design and practical condition first need to be optimised.