Pembina Pipelines Trust operates approximately 8,350 km of pipelines in British Columbia and Alberta. Pembina was awarded the contract to supply pipeline services to the new Oil Sands Project.
The pipeline project included the completion of five pipeline loops along the Alberta Oil Sands Pipeline (AOSPL) pipeline. Existing AOSPL right of ways (ROWs) were used wherever possible to limit the quantity of new disturbance caused by the pipeline project, which is preferred by government regulators. Construction of the Horizon Pipeline Project started in 2006. Construction management was handled by Sambrosa Engineering of Edmonton, Alberta.
One of the five new pipeline loops was the 17.3 km NPS24 Pine Creek Loop, which followed the existing pipeline corridor through the environmentally sensitive La Biche Wildland Provincial Park. There were many construction related challenges to deal with on this particular portion of the project. This loop involved the crossing of two water bodies, the La Biche River at the north boundary of the Park and Pine Creek at the south boundary. These water bodies constitute portions of the northern, eastern and southern boundaries of the park.
After 2002, new pipeline construction in the park was no longer permitted, as was demonstrated when another large diameter pipeline that was to be constructed at about the same time as the Pine Creek Loop was forced to reroute around the park because Pembina's existing ROW through the park was only licensed for a single pipeline. Pembina was given permission to construct the Pine Creek Loop within the Park because it had a permit for multiple pipelines within the AOSPL 30 metre ROW.
Article continues below…The original report was based on an isolated, trenched crossing design. However, this method of crossing was rejected by Alberta Tourism, Parks and Recreation based on the successful HDD crossing of the La Biche River at the north end of the La Biche River Wildland Park. Alberta Tourism requested and Fisheries and Oceans Canada subsequently permitted an HDD crossing method as a first attempt with the trenched method as last resort.
Worley Parsons (WP) began reviewing the various configurations for a directional drill and subsequently determined that at least some new ROW would be needed within the Park. Although summer construction would occur during the 16 April to 15 July restriction period, approval to proceed was obtained as a result of the change to a trenchless construction technique, which would not be required in stream work.
During the above process WP was planning for the Pine Creek HDD. Success of the HDD was contingent on obtaining adequate subsurface geotechnical information to assist in engineering the crossing; the geotechnical investigation was completed in January 2007. Construction of the Pine Creek Loop was carried out successfully, with the Pine Creek HDD slated for July 2007.
Due to the protected status of the La Biche River Wildland Park, no additional ROW was being granted within the park. WP attempted to design an HDD crossing with this restriction in mind but could not achieve the desired results based on the following:
- The existing AOSPL ROW configuration was designed to accommodate the previously installed trenched crossing. There were changes in horizontal direction both upstream and downstream of the crossing.
- A number of adjacent pipelines within the area had to be crossed along the HDD path.
- The minimum design radius of curvature, required clearance from the bottom of Pine Creek and the required entry and exit angles.
Ledcor Pipelines was awarded the mainline contract for the Horizon Pipeline Project. Ledcor subsequently awarded the HDD subcontract to Direct Horizontal of Stoney Plain Alberta. Direct completed the NPS 24 La Biche River crossing with an American Auger DD440 and moved the same drill rig to the south side of Pine Creek in July 2007. The selected drill rig (pictured) can be set up to enter at a minimum of 12 degrees, steeper than the designed entry of 10 degrees. Due to this change in the vertical alignment and the deepening of the bore profile as a result of steering difficulties, the total length of the drill ended up at 494.3 metres, an increase in length of about 12 percent.
Problems with fluid releases close to the rig were encountered as a result of the soft ground and the shallow depth due to the shallow entry angle. Ultimately the contractor chose to put in 40 metres of 42 inch casing and then installed 70 metres of 12 inch casing to minimise the fluid releases to surface. The 12 inch casing was utilised during the pilot hole drilling to try to ensure that the pilot hole was not lost as a result of the tripping in and out of the pilot hole in the soft clay formation.
In addition, the contractor requested changes to the drill path to allow for a deeper drill path to minimise fluid releases. The minimum radius for the drill was set at 600 metres at any point along the drill. Due to the soft nature of the clay, steering was somewhat difficult, requiring frequent adjustments. Ultimately the contractor was successful in completing the pilot hole and the first ream.
At this point the engineer was provided the as-built drill profile survey data for review. Engineering Technology Inc of Calgary (Entec) was contracted to review the survey data and provide information regarding the drilled radius of the crossing. Entec used the industry accepted ‘dogleg’ method to calculate the minimum radius along the drill path. The smallest radius found was 270 metres and another 16 points where the radius was below the minimum recommended radius of 600 metres.
The 270 metre radius was calculated to be close to failure under installation stresses and beyond failure under operating stresses. Due to the large discrepancy between the pilot hole minimum radius of 270 metre and the recommended minimum radius of 600 metres a more refined analysis of the crossing would be required to accurately determine if the current pilot hole drill path was in fact acceptable. There are a number of methods available for calculating the combined stress within an HDD drill path. The Tresca, Von Mises and limit states approaches are all recognised within the CSA Z662 and offer varying degrees of conservatism. Regardless of the method chosen stress is analysed for a pipeline within the HDD path for the maximum operational conditions.
The presence of applied loads in more than one direction results in a much more complex state of stress than for applied loads in only one direction (uniaxial). The predominant stresses in pipelines are typically biaxial, with internal pressure acting in the circumferential (hoop) direction and thermal loads and beam bending acting in the longitudinal direction. The yielding of the steel under these conditions is considerably more complex, and there are two widely accepted approaches for determining the combination of stress in various directions that result in yielding of ductile material (such as steel); the two approaches are the Tresca theory and the von Mises theory (Please refer to the conference paper for further information).
The combined stress as per CSA Z662 is a biaxial stress, which is a combination of hoop stress due to internal pressure and longitudinal stress due to bending of the pipe to achieve the curved shape of the HDD path added to the stress imposed on the pipeline due to thermal differential. A large positive thermal differential, together with internal pressure is typically the worst case.
Shortly after deciding to complete the 36 inch ream the project team recognised it may not be reasonable to expect the 36 inch ream, to improve the current minimum pilot hole radius from 270 metres to the minimum calculated radius of 425 metres as calculated in the Tresca analysis. This led to another refinement of the combined stress calculation. A more realistic approach is to compute combined stress as per the Von Mises formula, which uses similar inputs as Tresca with modest time and effort required. The Von Mises calculation method is also known to be more realistic for ductile steel such as employed in modern oil and gas pipeline systems.
For the Pine Creek crossing the soil within the drill path was assumed to be a ‘soft formation’ based on feedback from the drilling contractor, the difficulty experienced with down hole steering and the results of the geotechnical investigation carried out by AMEC Earth and Environmental. This resulted in the use of soil category A.
Analysis of the drill path was completed using the Von Mises method with the above parameters. The resultant minimum acceptable radius was determined to be 385 metres, a marginal improvement on the previous 425 metres. The 385 metre radius would need to be compared to the calculated minimum radius in the drill path after completion of the 36 inch ream.
The ream was completed two days later. Entec analysed the drill profile data and calculated a minimum dogleg radius of 284 metres, with a number of other radii less than 385 metres. The subsequent survey tool run was completed later in the same day. Entec analysed these results and calculated a minimum radius of 170 metres with a number of other radii less than 385 metres. The ever decreasing calculated minimum radius confirms that the progression of the calculations is from conservative to more realistic.
Pembina and its construction manager subsequently decided to pull the NPS24 pipe, as seen on page 55, into the hole and the project team agreed to continue analysing the available information to determine if in fact the 36 inch reamed drill path was acceptable.
First, the project team completed a sensitivity analysis to determine if reducing the MOP of the pipeline or increasing the pipe grade would be a viable alternative. After careful review, the required MOP reduction to 4,400 kPa or pipe grade increase to 545 MPa were still not sufficient to provide acceptability for the Pine Creek crossing.
The next and only viable alternative for analysing the crossing was to determine the absolute stress level within the pipe at this location. The Tresca and Von Mises methods are both stress based, conservatively not accounting for the nonlinear steel properties. Due to tight bend radii determined in the field, the group agreed to undertake a third method of analysis; a nonlinear strain analysis, which is permitted by CSA-Z662 Annex C as a limit states design approach. This strain analysis is much more complex than either stress method, but it also is a more rational and usually less restrictive approach. Determination of the longitudinal strain with nonlinear steel properties and subject to biaxial stresses is performed with a finite-element computer program.
Subsequent to the very detailed review of the stresses involved in this installation, the current drill alignment was supported by Worley Parsons Calgary, based on the maximum calculated strain being below the maximum allowable strain for the crossing using limit states analysis. Colt used the minimum drill path radius of 170 metres and compared the strain at this location to the maximum strain allowable for the NPS24 crossing pipe. The results were deemed acceptable as per the above discussion.
In addition, Colt has recommended running a geo-pig through the crossing within two years of the start of pipeline operation. This can be used to verify the geometry of the pipe within the crossing and reconfirm the results of the limit states analysis. It is anticipated that the final geometry of the drill profile will be better than the lowest indicated radius of 172 metres.
The purpose of this article is not to support less conservative analysis as a general rule for HDD installations but rather, given the difficult circumstances encountered at the site, to illustrate that this method of analysis is applicable and required in certain circumstances. HDD installations of pipelines beneath pipeline route obstructions are technically challenging, as are the determination of the construction stresses. These installations are such that future repair and cleanup if a problem develops, are not possible or are extremely difficult due to the location and depth of these installations. Therefore, a conservative approach to the design provides some assurance that problems should not occur at this type of crossing.
Basket is empty.
