Non-Destructive Pipe Wall Thickness Assessment

Echologics Engineering Inc. has moved ahead with a collaborative research agreement with the National Research Council of Canada (NRC) to further develop and commercialize NRC’s patented pipe wall thickness assessment technology. The new technology measures the remaining wall thickness of buried water transmission and distribution pipes in a non-destructive way and without taking pipes out of service by using acoustic signals. Echologics also signed a joint project agreement with the City of Hamilton, Ontario/Canada, to evaluate the technology by undertaking extensive pilot tests on pipes in the City’s water distribution network.

Metallic pipes lose thickness as a result of internal and external corrosion. Cement-based pipe types lose “effective” wall thickness by the weakening of the wall as a result of the leaching out of the cement by aggressive waters. Pre-stressed concrete cylinder pipes also lose strength as a result of the corrosion of pre-stressing steel. Plastic pipes may lose material strength as a result of mechanical fatigue and chemical attack. A direct indicator of the general structural condition of pipes is the remaining or residual wall thickness. Also, if wall thickness loss is monitored over time, the remaining service life of pipes can be estimated.

Existing methods to estimate the pipe wall thickness, such as pipe sampling programs and the remote field eddy current technique, are too disruptive and costly to be justified by water utilities for inspection at the network level. The NRC technology provides a very effective solution.

Pilot tests

Pilot tests using the new technology for measuring wall thickness were performed at 11 sites in the City of Hamilton. Test sites were chosen in different parts of the City based on soil type and break history of the pipes. They included large and small diameter pipes at different levels of deterioration (two pipes had a diameter of 20 inches). The pipes were all of the cast iron type installed between 1860 and 1960 and included both pit-cast and spun-cast pipes.

The accuracy of the remaining pipe wall thickness measured by the new technology was evaluated by comparing it with the average wall thickness and visual appearance of pipe samples exhumed at 8 of the test locations. A one-metre long pipe sample was exhumed at each location by PipeFlo Inc. under contract with the City of Hamilton and the supervision of the City’s consultants for this project, UMA Engineering Group Limited. The samples were sand blasted and analysed to determine average wall thickness, average pit depth, % pitted area and a variety of other corrosion-related information. Correng Consulting Services did the analysis under a sub-contract from UMA.

Results

Pipe wall thicknesses that were measured by using the NRC method were in excellent agreement with average thicknesses and visual appearance of exhumed pipe samples reported by Correng, with few minor exceptions. The most dramatic and interesting agreement was for the pipe at one particular site. Predicted thickness loss for two pipe sections at this site were 51 and 33%, which were significantly higher than losses predicted for pipes at other sites. Corrosion analysis by Correng indicated that pipe samples from this particular site were in extremely poor condition in comparison to samples from other sites. These samples exhibited severe external corrosion that resulted in numerous perforations. In addition, they showed very poor casting with numerous large voids. Several photographs of the sandblasted pipe sample from this particular location are shown here (click on photos to enlarge).

How does the new technology work?

The new technology works by measuring how quickly acoustic signals are transmitted along a section of pipe. Acoustic signals are induced in pipes by releasing water at fire hydrants in a controlled manner. Then, they are measured using acoustic sensors positioned at two longitudinally separated points on a pipe. The sensors are attached at easy-to-access points, such as fire hydrants and control valves, or directly on pipes in existing access manholes. A schematic of the measurement setup is shown below (click on it to enlarge). The acoustic propagation velocity is calculated based on the sensor spacing and time delay between the measured acoustic signals. Average wall thickness of the pipe section between the acoustic sensors is then back calculated from a theoretical model of its relationship with the acoustic velocity, the pipe’s internal diameter and Young’s modulus of its wall, and the bulk modulus of elasticity of water, all of which are usually known or easily determined.

The length of the pipe section over which the acoustic velocity is measured can be arbitrarily chosen. Initially, a 100 to 200 metres long section, which is the usual distance between fire hydrants or valves in urban areas, may be chosen. Subsequently, if a higher thickness resolution is needed, for example, when a bad section of pipe is found, or when the client has concerns about a particular section, the resolution can be increased by moving the acoustic sensors closer together. To do so, closely spaced small holes to access the pipe may be drilled, for example, by using keyhole vacuum excavation equipment. Alternatively, arrays of closely spaced hydrophones may be inserted into pipes thru corporation stops or fire hydrants.

Velocity measurement can be performed with hardware normally used for locating pipe leaks using the cross-correlation method. However, measurement of the velocity tends to be more technical than the usually straightforward leak correlation. Velocity measurement and wall thickness calculations are made in real time using specially developed software, trademarked as ThicknessfinderRT. Recent research and development have led to several improvements of the technology that include a refined theoretical model for non-uniform pipe sections, an optimal procedure for acoustic velocity measurement, and a method for inspecting the quality of the measurements.

The pipe wall thickness determined by the new technology represents an average value for the pipe section over which the acoustic velocity is measured. This is not a limiting aspect of the method. Generally, pipes will have a more-or-less uniform thickness profile over significant lengths, say 50 to 100 metres, as soil and bedding conditions are unlikely to change significantly over such distances. Also, average wall thickness values are the most suitable to evaluate the residual life of pipes for the purpose of long-term planning of rehab and replacement needs.

The new technology can be used for all types of pipes, including cast and ductile iron, steel, PVC, asbestos cement and pre-stressed concrete cylinder pipe (PCCP).

Application of the new technology

The new technology is anticipated to be a powerful additional tool for the City of Hamilton’s asset managers to prioritize which pipes in the City require replacement or rehabilitation. An added bonus from the City’s perspective is that leak detection can be done at the same time as the pipe wall thickness measurement (both are done using the same measurement setup but one with ambient acoustic noise and the other with an active acoustic source). Mr. Kevin Bainbridge and his team at the City of Hamilton’s Capital Planning and Implementation Division are working with from UMA Engineers to determine the best way to utilize the results of the measurements. Mr. Bainbridge anticipates that the City of Hamilton will begin to use the technology in the spring of 2006 to help assess their water mains.

Market potential

Cast and ductile iron pipes comprise 70 to 80% of most water transmission and distribution networks in Canada and the United States. The remaining part is comprised of non-metallic pipes, e.g., asbestos cement, plastic (PVC or PE), and / or pre-stressed concrete cylinder pipes (PCCP). A large proportion of metallic pipes are fast approaching the end of their expected life. The replacement cost faced by water utilities will be huge. The American Waterworks Association estimates replacement costs for water pipes in the United States to be US $6,300 per household, which is most likely true also for Canadian systems [Reference: Dawn of the Replacement Era - Reinvesting in Drinking Water Infrastructure. Published in 2002 by the American Water Works Association, Denver, CO., available on the Internet at: www.awwa.org/Advocacy/govtaff/infrastructure.pdf (cited 28 April 2005).]. According to this, the total replacement cost for medium-sized cities like Ottawa or Calgary is in excess of 1 billion dollars. Replacement costs are expected to peak in about 20 years when water pipes installed during the construction booms of the 1920s and 1930s and post World War II period begin to fail en masse.

Traditionally, decisions to replace or rehab buried pipes have been based on general indicators such as failure history, age, type, and size of pipes. Such decisions do not lead to the most efficient use of the limited financial resources of municipalities. Optimal decisions require reliable information about the actual condition of buried pipes. Gaining access to the pipes to inspect them is difficult, disruptive, and very costly. In many situations also, pipes cannot be taken out of service to be inspected. Therefore, condition assessment has to be done in a non-destructive and non-intrusive way.

One approach to obtain information about pipe wall thickness is to exhume small samples from pipes. Some water companies in the U.K. have adopted this approach; for example, Yorkshire Water takes a 300 mm long sample every kilometre of pipe. The sample is cut in half and sandblasted to measure the remaining thickness. The remaining service life of pipes is then determined based on statistical analysis. Another approach is to measure the wall thickness in situ by using ultrasonic or remote-field eddy current techniques. The required instrumentation for these techniques is set up in remotely controlled vehicles (known as pigs), which are launched inside pipes. Unlike the sampling approach, these techniques provide a continuous profile of the wall thickness.

The physical samples approach is costly and disruptive. Also, thickness values based on exhumed samples may not be representative of the overall condition of the pipe unless a very large number of locations are excavated. Launching of testing vehicles inside pipes is a very complex operation. It requires taking the pipe out of service, scraping and sweeping the pipes to remove tubercles and debris, and the construction of pig launching stations. Also, related data acquisition and analysis are very intensive. Therefore, the cost and level of disruption of deploying testing vehicles are too high to be justified by water utilities. ItÍs believed that because of this, Hydroscope Inc., a Canadian company that has developed and marketed the successful remote field technology, went out of business.

NRC’s new technology for assessment of pipe wall thickness overcomes the limitations of existing technologies. ItÍs cost-effective, accurate, efficient, easy to implement, non-destructive and non-intrusive. This will help to open the door widely to the huge North American market of more than 50,000 water distribution systems.

More information

For more information and / or to arrange for a trial of the new technology, please contact Marc Bracken at Echologics Engineering Inc. [tel.: +1 (416) 249-6124, fax: +1 (416) 249-8833, e-mail: marc@echologics.com.