Deterioration due to age, urban growth, and pipe material quality has necessitated the need for replacing existing sewer and water mains. Pipe bursting is rapidly gaining acceptance as an alternative method for renewing existing sewer, water infrastructure systems and even city gas distribution pipelines. Pipe bursting enables the replacement of an existing pipe with one of equal or larger diameter along the existing trajectory with minimal disruption to surface activities.

Overview of pipe bursting methods

Static Method

Static methods burst the original pipe using static forces, or forces developed from the geometry of the bursting head as it is pulled or pushed through the existing pipe. A pulling force is applied to the cone shaped bursting head through rods, cables, or chains. The bursting head then is pulled through the pipe causing the existing in ground pipe to fail in tension by the radial force applied to the pipe wall from the cone within the pipe. As the host pipe is burst, the bursting head pushes the broken pipe pieces into the soil as it displaces the surrounding soil, thus creating a cavity for the new product pipe. Both continuous and sectional pipe can be installed using static pipe bursting.

For the installation of continuous pipe, such as high-density polyethylene, ductile iron, polyvinyl chloride (PVC), or steel, access pits must be excavated at each end of the pipeline to be replaced. On one end of the line, the machine pit is excavated into which the pipe bursting machine that pulls or directs the bursting head is located. Opposite the machine pit is the insertion pit through which the new pipe or product pipe and bursting head are inserted into the existing or host pipe. The setup for a typical burst using static pipe bursting is shown in Figure 1. Any services along the pipe route connected to the original pipe must be disconnected prior to the start of the burst with access to the connections achieved through service pits.

A slightly different setup is required if sectional non-welded joint (i.e. clay, ductile iron, fibreglass, polyvinyl chloride, or reinforced concrete) pipe is used to replace the existing line including. Again, access pits are excavated at each end of the line to be replaced, except in this case both pits are considered machine pits. The installation of sectional pipe requires that constant force be applied to the pipe to keep the joints together during installation. This may be achieved by using a chain or cable run through the new line from the bursting head to a trailing plate on the last pipe section, or alternatively by using a push-pull technique. In the push-pull setup, the bursting head would be pulled by one machine in the pulling pit, while in the opposite pit; the pipe section would be pushed by another machine as illustrated in Figure 2. In this setup, a constant pressure is applied to the new pipe during installation by maintaining the push force slightly higher than the pulling force. This requires the synchronisation of the two machine forces while allowing large diameter installations to be achieved. One such installation occurred in St. Petersburg, Florida, where 230 m (770 ft) of 900 mm (36 inch) diameter vitrified clay pipe was successfully replaced with 900 mm diameter Hobas pipe (Thomas, 1996).

Pneumatic Method

The pneumatic pipe bursting method illustrated in Figure 3 utilises a bursting head that displaces the soil using a horizontal hammering force developed with air from a compressed air system. Using compressed air, the bursting head is able to develop a hammering rate of 180 to 580 blows per minute. The cone shaped bursting head is driven through the soil like a nail being driven into a wall. Each blow impacted by the bursting head into the pipe creates an impact load, applying a ‘hoop’ stress into the pipe causing it to burst in tension. In addition, the hammering action creates force in the longitudinal orientation, causing failure in shear as the pipe is ripped. The shape of the bursting head combined with the percussive action push the pipe fragments into the soil providing the space necessary for the installation of the product pipe.

With this method of pipe bursting, the bursting head is guided through the pipe with the use of a tensional cable inserted through the pipe prior to bursting. This cable is attached to the bursting head and provides constant pulling tension, through the use of a winch, to keep the bursting head in contact with the host pipe and aligned with its path, as well as assist in pulling the new host pipe into place. The main force that allows the progression of the bursting head through the pipe comes from the percussive hammering action of the pneumatic head itself. Both the air compressor and the winch are set at constant pressure and tension that allows the operation to proceed with little operator intervention until the pipe section is burst. To power the bursting head, compressed air lines (hoses) must be run through the new product pipe.

Hydraulic Method

This method of pipe bursting is defined by the method in which the host pipe is burst. Rather than the pipe being burst from the transfer of an axial pulling or hammering force radial into the plane of the pipe diameter, the bursting head expands, radially fragmenting the pipe from inside. Using hydraulic cylinders, the head expands to burst the pipe, then contracts to allow the winch to pull the cable and advance the head incrementally forward. The winch or pull on the cable does not assist in the bursting of the pipe, but rather pulls the head to help displace any residual soil formation as well as pull the product pipe into the expanded cavity.

Similar to the pneumatic pipe bursting system, the hydraulically expanding bursting head requires a power source to provide energy to burst the pipe. In this case, a portable power unit at the machine pit on the surface provides power for the hydraulic cylinders, with hydraulic hoses run inside the entire length of the product pipe. The hydraulic pipe bursting process is illustrated in Figure 3.

Pipe Bursting vs Open-Cut

In comparison to conventional open-cut techniques of pipe replacement, pipe bursting has several inherent advantages. From experience, it has been shown that in almost all circumstances, pipe bursting has cost less that open-cut alternatives, completed the installation in less time and, as a result, been a more efficient construction method. The main advantage pipe bursting has over open-cut methods is that in pipe bursting a minimal amount of excavation is required. This especially comes into play when one compares the cost of open-cut excavation to that of pipe bursting. As the depth of installation increases, the cost of installation using pipe bursting almost remains constant. In comparison, the cost of open-cut excavation increases greatly as the depth of the installation increase. This is largely due to the increased need for dewatering, shoring and extra excavation often required during open excavation operations (Poole et al., 1985).

There are also social costs to consider when one compares pipe bursting to open-cut excavation. These include the intangibles that one cannot accurately quantify in financial terms. These costs include the loss of access to a driveway or garage, or the closing of sidewalks and parking areas in a downtown business district. This is where pipe bursting has a distinct advantage over open-cut, especially in urban centres or non-green field applications (McKim, 1997). To replace a line using open-cut methods, the entire line must be excavated while with pipe bursting, only a machine pit and insertion pit need be excavated. This minimises the interference with street traffic, reduces noise pollution, reduces environmental disturbance and reduces ground settlement, as there is minimal pore pressure reduction in the soil strata. A typical pipe bursting replacement can be performed using only a one-lane right of way, thereby maintaining the flow of traffic on the road.

In general, the selection of pipe bursting as the replacement option can be attributable to situations where restoration costs are high, such as beneath roads or highly landscaped areas, or in areas of high underground utility congestion. Additionally, if the authorities prohibit the use of open-cut methods in environmentally sensitive areas, or in locations of high traffic, pipe bursting becomes a favourable replacement method. Perhaps the situation where pipe bursting becomes the most viable, if not only option for replacement, would be when the pipeline right-of-way is inaccessible due to existing structures or obstructions, and utilities. These situations make the utilisation of pipe bursting more feasible than conventional cut and cover options.

The main contrasts between trenchless and open-cut projects are presented in Table 1. The table compares several planning elements that are common for the rehabilitation or installation of subsurface utilities. In most cases, due to the reduction of excavation and spoil handling, site restoration, and speed of the operation, pipe bursting methods can complete the rehabilitation of pipes in a shorter time and with reduced social costs compared to traditional open-cut methods. The issue of traffic control is of great concern when working in areas of high vehicular and pedestrian usage. In many situations, the closing of a route or street due to construction activities causes disruption to the businesses and residences in and around the closed road. Loss of access for customers, increased traffic due to detours and noise are the effects of construction activities.

Through the utilisation of trenchless construction methods, many of these problems can be minimised or eliminated. One of the main advantages of pipe bursting in the replacement of pipes beneath roads is that only one lane is required to situate the equipment necessary to perform the replacement. Subsequently, traffic flows can be maintained along the site in question minimising the impact on the surrounding area.

Most trenchless pipe replacement techniques generally require some amount of excavation to be performed to complete the installation. The amount of excavation required on a pipe bursting project is substantially less than that required on an open-cut project for the replacement of the same pipe or conduit. This translates into reduced risk due to minimising the time required for crews to work in excavations, as well as minimising the ground disruption around buried utilities and building foundations (Lueke and Ariaratnam, 2001).

In comparison to other trenchless methods, the decision to use pipe bursting over another rehabilitation or new construction method is determined by the present and future needs of the infrastructure system. In situations where the volumetric capacity of the line needs to be increased, pipe bursting provides the only trenchless alternative (Ariaratnam et al., 2001). In comparison to conventional lining methods, where the original pipe must be structurally sound to provide an effective rehabilitation, pipe bursting provides a complete structural replacement independent of the original condition of the line.

Additionally, if a new line or grade is required, pipe bursting is incapable of changing these conditions and therefore directional drilling, auger boring, microtunnelling or pipe jacking would be better trenchless alternatives. Lastly, consideration must be given as to whether the existing pipe requires a complete replacement, complete or partial lining, or only spot repairs to fulfil the existing and future service requirements. These considerations determine how pipe bursting compares to other trenchless rehabilitation and new construction methods for a particular application.

Engineering considerations

Engineering considerations are paramount when undertaking a pipe bursting project. Project planning, geotechnical conditions, pipe upsize, depth of cover, bursting head design, and strain monitoring are all issues that must be considered and addressed prior to installations using any pipe bursting method (Ariaratnam and Chan, 2005). Proper investigation of all of these factors is imperative to minimising risk and ensuring project success.

Project planning

In the planning of a trenchless replacement project, consideration must be given to the arrangement of the pulling machine and pipe insertion pits, since the most time consuming operation in the pipe bursting process is the setup of the machine and the excavation of pits. Therefore, the number of setups and amount of excavation should be minimised. Prior to mobilisation, the planner must consider the arrangement of pipe sections and manholes. It is best to plan multiple uses for machine pits to minimise equipment transportation and relocation, as well as the number of times the power plant is required to be setup. If a section of pipe that is being burst is a continuation or is in line with another section of pipe joined with a manhole, that manhole would best serve as a machine pit. With this arrangement, after the first section of pipe is burst, rather than the rods or chain being disconnected as they are pulled, the rods can be shunted (or chain pulled) down through the next section of pipe to be burst. This increases the productivity of the operation by shunting rods (or chains) down one section of pipe while simultaneously bursting another.

Geotechnical considerations

Similar to other underground construction techniques, geotechnical conditions are important considerations in determining the feasibility of utilising pipe bursting on a particular project. In general, soils that are best suited for open-cut excavation are also well suited for pipe bursting. Problems may occur when bursting in dry granular soils such as sand. In particular, as the burst progresses, vibration from the operation can assist in the compaction and constriction of the sand around the product line. This redistribution of the sand may increase the friction caused by the soil, therefore increasing the force required to pull the new product line in place. Regardless of the type of soil around the pipe, the amount of force required to pull the product line increases with the length of installation. Subsequently, if the sand has constricted around the pipe, as the length of pull increases, the force necessary to complete the pull may exceed the pull or push capabilities of the pipe bursting equipment being used on the project. This emphasises the importance of proper site investigation to determine the composition of the soil around the existing pipe and consequently to ensure that properly selected equipment is used to complete the installation. Figure 6 illustrates various geological conditions in order of increasing difficulty.

Pipe upsize

During pipe bursting, a volumetric displacement of the fragmented pipe and ground material will occur. It is important to determine the extent of this displacement for several reasons. Firstly, pullback force is proportional to volume and surface area – that is, soil shearing resistance is proportional to volume sheared. Secondly, surface heave is also proportional to volume displaced. Figure 7 illustrates upsizing consideration categories of: 1) easy (< 30 per cent volume increase); 2) moderate difficulty (30 to 100 per cent volume increase); and 3) very challenging (> 100 per cent volume increase) conditions based on volumetric displacement.

With proper locating and identification of the utility right of ways around the host pipe and pre-construction surveys along the right-of-way, measures can be taken to reduce the effects of the ground movements on these utilities. To reduce the effects of surface heave, it may be possible to place additional load on the surface in areas of concern while the burst is conducted (Lueke and Ariaratnam 2000). The additional weight on the surface may assist in redistributing the ground displacement more in a lateral direction, therefore reducing the surface heave.

Depth of cover

Perhaps the greatest impediment in the adoption of pipe bursting to main line pipe replacement is the uncertainty associated with ground movements. Regardless of whether the original pipe is replaced size for size, or with a larger pipe, the ground around the original pipe cavity will ultimately move. Therefore, care must be exercised whenever bursting in close proximity to buried utilities and conduits. Ground movements may also occur on the surface if the depth of cover above the pipe being burst is too shallow. The minimum depth of cover is dependent on soil conditions, installation geometry, size of host pipe, and degree of upsize. The vertical ground movement will manifest itself as surface heave. Depending on the nature of the cover material, this surface heave may cause permanent damage to pavements or foundations above the bursting operation. Therefore, it may be prudent to monitor existing structures, utilities, and pavement during installation. Upheave of ground surface may occur if the soil density and depth of cover is inadequate to accommodate the volumetric expansion created by the normally larger bursting head and/or increase in diameter of the new replacement pipe during an upsizing operation.

Bursting head design

Bursting head design is an important engineering consideration for the success of a pipe bursting project. Numerous external forces affect the overall bursting process as illustrated in Figure 8. Subsequently, there are a number of factors to consider when designing a suitable bursting head for the anticipated conditions. During a pipe replacement operation, a bursting force or the force required to break the existing pipe into small pieces is encountered. The magnitude of this force is dependent on several factors including the pipe material strength and behaviour and the pipe size and wall thickness.

The material strength of the existing pipe is important for an obvious reason. The stronger the material, the more force required to break it. While not quite as evident, the material behaviour of the existing pipe, whether brittle or ductile, also has an important affect on the bursting force. When the bursting head travels through a brittle pipe, the existing pipe fails to fracture into many small pieces by exceeding the maximum hoop stress. Were the pipe not contained by the surrounding soil, these fragments would explode outward. In a ductile pipe however, the existing pipe usually will not fail without some starting point of initial failure. As such, cutting fins are often welded to the bursting head prior to a ductile pipe replacement pull. These fins apply point loads on the inside perimeter of the existing pipe, inducing it to fail in a similar manner as for brittle pipes. This similarity is correct to the extent that the ductile pipe fails by overcoming the maximum hoop stress resistance, but false in that the ductile pipe does not break into many small pieces, but rather tears into strips equal in number to the number of cutting fins attached to the bursting head. Therefore, the material strength affects the force required to break the pipe while the material behaviour determines the number of times it is broken.

The next two influencing factors, pipe size and wall thickness, are more relevant for brittle pipes than for ductile ones. From past industry experience, the larger the brittle pipe, the more pieces it breaks into. For a ductile pipe, the larger size has no direct implication on the number of pieces it breaks into, except that more cutting fins are usually added the larger the pipe, again resulting in more pieces. The existing pipe wall thickness is important because it affects the amount of material strength that is mobilised before failure. Thinner walled pipes break sooner than thick walled ones. Not quite as obvious is the fact that thicker walled brittle pipes also break into larger pieces, which affects the breaking length, defined as the length parallel to the direction of the pipe pull, again resulting in larger bursting forces. In contrast, since the failure mechanism in ductile pipes is more of a tearing than fracturing behaviour, this breaking length is not as critical; the tearing action is an almost continuously occurring process.

Bursting head design should be based on upsize volume, pipe material, depth of cover, and length of run. Figure 9 illustrates a bursting head with an added cutting fin for replacing an existing cast iron water main. Pipe with ductile failure such as PVC, steel, and high density polyethylene require the addition of a cutting device to slit the material while simultaneously performing the burst. For pipe with brittle failure mechanisms such as concrete, asbestos-cement, and clay a barrel type bursting head is preferred as illustrated in Figure 10.

Strain monitoring

Most manufacturers of high density polyethylene pipe will only guarantee pipe material integrity for strains of less than 5 per cent during installation. A program to monitor strain during pipe bursting was conducted on the Millstone Sanitary Trunk Sewer Replacement project in Nanaimo, BC, Canada (Ariaratnam and Harper, 2002; Ariaratnam et al., 2001). A combination of strain gauges and linear potentiometers were installed within the pipe approximately 4 m behind the bursting head. The strain gauges were used to measure localised axial movement, while the linear potentiometers were used to measure strain over a 1 m distance. In total, four strain gauges and four linear potentiometers were installed in combination at quarter points inside the pipe. Initially, the instrumentation was attached by a silicon sealant only; however, the linear potentiometers became unattached from the pipe wall during the installation. The strain gauges remained attached and functioned as expected. As a result of the initial trial, it became apparent that the instrumentation needed to be attached in a more secure fashion. This was accomplished by attaching the linear potentiometers to the inside pipe wall using metal screws. The metal screws caused negligible damage to the pipe wall due to its sufficient thickness. In total, five replacement sections had strain measurements conducted at various locations throughout the project. Four of these involved strain measurements being recorded at a distance of 4 m behind the bursting head, while the last instrumented section had strain measurements recorded at the mid-section in addition to directly behind the bursting head. The maximum strain exhibited was less than 1.5 per cent, which is significantly less than the 5 per cent ceiling. An example of strain measurement results is illustrated in Figure 11.

Conclusions and recommendations

With ageing underground infrastructure and increased urban growth, the utilisation of trenchless methods such as pipe bursting will continue to rise. As municipal engineers look towards adopting innovative technologies for replacing sewer and water main while providing reduced costs, increased productivity, and minimised disruption to surface activities, pipe bursting projects are expected to increase significantly worldwide. In comparison to open-cut trenching solutions, the unique nature of pipe bursting requires considerations that are unlike conventional replacement options. A major advantage that pipe bursting presents over other methods of pipeline renewal is its capability to increase flow capacity through upsizing, and install a new line that functions independently of the original pipe. The technique can be used to replace pipes of various composition and sizes, with new pipe materials that can be of either sectional or continuous nature, in both main line and lateral replacement applications. Project planning, geotechnical conditions, pipe upsize, depth of cover, bursting head design, and strain monitoring are all engineering considerations when engaging in pipe bursting. Lessons learned from projects that require engineered solutions are invaluable in developing the technology to provide new applications in the trenchless renewal arena. Through case histories and discussion with the industry, the advantages of the application of trenchless pipe replacement as a viable alternative to conventional pipe installation and rehabilitation methods will be more recognised.

References

American Society of Civil Engineers (2005). Infrastructure Report Card. ASCE, Reston VA.

Ariaratnam, S.T., and Chan, W. (2005). ‘Engineering of Trenchless Pipe Renewal Projects for Infrastructure Renewal’ Proceedings of the 1st CSCE Conference on Infrastructure Technologies, Management and Policy, CSCE, Toronto, Ontario, June 2-4, FR106(1) – FR106(10).

Ariaratnam, S.T., and Harper, R. (2002). ‘Pull Load Evaluation of Static Pipe Bursting’ Proceedings of the Underground Construction Technology ‘2002 Conference, Houston, Texas, January 15-17 (on CD).

Ariaratnam, S.T., Harper, R., and Cyre, G. (2001). ‘Monitoring and Instrumentation of Trenchless Technology Projects’ Proceedings of the 2001 International Conference on Underground Infrastructure Research, Balkema Publishing, Kitchener, Ontario, June 10-13, 225-230.

Harper, R., Sims, W., and Ariaratnam, S.T. (2000). ‘Pipe Bursting of the Millstone Sanitary Trunk Sewer’ Proceedings of the CSCE 28th Annual Conference, London, Ontario, June 7-10, 35-40.

Lueke, J.S., and Ariaratnam, S.T. (2001). ‘Rehabilitation of Underground Infrastructure Utilizing Trenchless Pipe Replacement’ Practice Periodical on Structural Design and Construction, ASCE, 6, 2, 25-34.

Lueke, J.S., and Ariaratnam, S.T. (2000). ‘Subsurface Ground Movements Associated with Trenchless Pipe Replacement Methods’ Proceedings of the 6th ASCE Construction Congress, Orlando, Florida, February 20-22, 778-787.

McKim, R.A. (1997). ‘Bidding Strategies for Conventional and Trenchless Technologies Considering Social Costs’ Canadian Journal of Civil Engineering, CSCE, 24, 819-827.

Poole, A.G., Rosbrook, P.B. and Reynolds, J.H. (1985). ‘Replacement of Small Diameter Pipes by Pipe Bursting’ Proceedings of No-Dig International ‘85, London, United Kingdom, 147-159.

Thomas, A. (1996) ‘Push-Pull Pipe Bursting Restores Sewer at Thunderdome’ Trenchless Technology Magazine, Peninsula, Ohio, September, 36.