Surface Laid Pipelines, as the name suggests, are the Pipelines which are laid on the surface of the loose desert sand without any restraints. These unrestrained pipelines, unlike the other unrestrained pipelines, do not have pipe supports to restrain the displacements, hence different approach and technique is implemented while analyzing such lines.\n\n\n\nThe purpose of this article is to illustrate the method to model and analyze surface laid pipelines using Caesar II and the considerations to be taken care of during the Stress Analysis.\n\n\n\nThe surface laying of the pipelines is opted by the client for Cost-benefit. It saves construction costs, maintenance costs, and operational costs largely.\n\n\n\nSites for Surface Laid Pipelines\n\n\n\nSurface laid pipelines have basically opted for the desert sites, where the pipelines are laid in the corridor from the Wellhead up to the Gathering Facility.\n\n\n\nAbove Ground Surface Laid Pipelines\n\n\n\nThe above-ground surface laid pipelines (Fig. 1) shall be routed in a manner such that no excessive movement occurs on the pipes due to the effects of thermal expansion and\/or contraction, internal pressure and other design internal or external loads. The axial and lateral expansions of above-ground pipelines shall be limited as far as possible. Expansion loops to be designed to accommodate the axial movements based on CAESAR II stress analysis recommendations for above ground pipelines.\n\n\n\nThe above-ground pipeline shall be modeled as "Unrestrained" in the CAESAR II software and should be analyzed using ASME B31.4\/B31.8 code calculations.\n\n\n\nThe unsupported surface laid pipelines shall be installed as unrestrained pipeline & shall follow natural grade elevation\n\n\n\nFig. 1 and Fig. 2 shows typical site images explaining how surface laid pipelines would look like.\n\n\n\nFig. 1: Typical Surface Laid Pipeline in Desert Area\n\n\n\nConsiderable Points for Surface Laid Pipelines\n\n\n\nThe below-mentioned points must be considered while designing Surface Laid Pipelines.\n\n\n\nWall Thinning of Hot BendsWall Thickness at Road CrossingsMinimum Radius of Elastic and Cold Field BendsLongitudinal Stress, Anchor Force and Free End Expansion CalculationEquivalent\/Longitudinal Stress CalculationUpheaval BucklingStability\n\n\n\nFig. 2: Site Photo of Surface Laid Pipelines\n\n\n\nWALL THINNING OF HOT BENDS:\n\n\n\nThe process used for bending a hot bend results in some thinning of the pipe wall thickness. An indication of this thinning as a percentage of wall thickness may be determined by the empirical formula stated in BS 8010-2.8 para 188.8.131.52\n\n\n\n\n\n\n\nn= is the inner bends radius divided by pipe diameter\n\n\n\n2. WALL THICKNESS AT ROAD CROSSINGS:\n\n\n\nThis calculation determined the minimum wall thickness required at road crossings to withstand normal vehicular loads. The calculation will be based on the method used in API RP 1102.\n\n\n\n3. MINIMUM RADIUS OF ELASTIC AND COLD FIELD BENDS:\n\n\n\nFor liquid lines, the minimum radius of curvature for cold field bends is determined as per the requirement of ASME B31.4 clause 404.2.2 to ensure that the bend will not be overstressed. For gas lines, the minimum radius of curvature for cold field bends is determined as per the requirement of ASME B31.8 table 841.2.3-1 to ensure that the bend will not be overstressed.\n\n\n\n4. LONGITUDINAL STRESS, ANCHOR FORCE AND FREE END EXPANSION CALCULATION\n\n\n\nThe aim of this calculation is to determine the longitudinal stress induced in a restrained \/ unrestrained pipeline as a result of the difference between the installation temperature and the operating pressure and also to calculate the potential anchor force; free end expansion and the virtual anchor length.\n\n\n\nASME B31.8 states that sum of the expansion stress, longitudinal pressure stress and applied bending stress must not exceed the 90% specified minimum yield strength of the material. In addition, the longitudinal pressure stress and the applied bending stress must not exceed 75% of the SMYS for the gas pipelines.\n\n\n\n5. EQUIVALENT\/LONGITUDINAL STRESS CALCULATIONS:\n\n\n\nASME B31.4 uses a term called the Equivalent Tensile Stress to control the combined stresses in the liquid pipelines. The Equivalent Tensile Stress is equal to the sum of the hoop stress and the longitudinal stress imposed on the pipeline. The limit for equivalent tensile stress is set at 90% of SMYS.\n\n\n\n6. UPHEAVAL BUCKLING:\n\n\n\nAn upheaval buckling calculation may be performed to ensure that the longitudinal forces are not sufficient to force the pipelines out of the ground. The required downforce will be calculated using the method described in OTC 6335 - Design of Submarine Pipelines against Upheaval Buckling, 1990. The required downforce is dependent on the axial force, which is calculated in the Thermal Expansion Calculation Sheet.\n\n\n\n7. STABILITY:\n\n\n\nThe method used to determine the vertical stability of the pipeline through water courses is to calculate the total weight of the pipeline and compare it with the weight of the external fluid (water) it would displace. In order for the pipeline to be considered stable, the pipeline weight must be 10% greater than the weight of the water displaced\n\n\n\nMajor points during Stress Analysis\n\n\n\nDuring the stress analysis, the analyst shall check the maximum lateral deflections in the system to make sure that the adjacent flow line does not cross each other while in operation considering the spacing between the flowlines mentioned in the specifications.\n\n\n\n 2. The design shall be adopted to add flexibility by introducing offsets\/expansion loops across the route while the lateral deflections of the line shall be within the limits so that the group of lines will not interfere with adjacent lines. \n\n\n\nProcedure for modeling in CAESAR II\n\n\n\nThe methodology for modelling the Surface Laid flowlines are bit tricky and tedious, so proper care has to be taken while modelling the same in the CAESAR II. Since the entire system is without restraint, a slight error might give improper results.\n\n\n\nBelow is the basic methodology for modelling the Surface laid flowline.\n\n\n\nSince the pipeline is surface laid (except at road\/track\/ other pipelines crossings where it is U\/G), a direct method for modelling surface laid lines is not available in CAESAR II (or any other such software) due to the continuous supporting (equivalent to say infinite number of supports) by natural soil along pipeline routes. The continuous soil supports are therefore taken care of by discrete soil springs and also by selecting such close springs at a distance not more than 5D (i.e 5*219.1 = 1095.5 mm as an example for 8\u201d). Support spacing shall be taken to nearest round figure (say for 8\u201d as an example at every 1000 mm), using the approximate stiffness in the vertical direction and also using friction value of 0.5 (which varies with location for such modelling). \n\n\n\nThis modelling approach has been recommended by COADE (Developer of CAESAR II). \n\n\n\nFig. 3: Surface Laid Pipeline Modeling\n\n\n\nThe CAESAR II procedure for finding the soil support stiffness is as given below:\n\n\n\nA small stress model of the pipeline shall be created using the actual design data. The small model could be with 25 nodes, with each node spaced at the 5D distance of the line. The model shall be buried using the buried modeler of CAESAR II \u00ae, with an assumed soil depth of half the diameter of the pipeline and using the typical soil design parameters. Upon buried conversion, the program inserts soil springs in the converted model of the sample model. The Y2 stiffness shall be read as seen in the converted sample model corresponding to a mid-node number (made free from the end conditions). This is the stiffness to be used in the actual pipeline model as normal rest support.\n\n\n\nIn case the stress model with 5D support spacing creates convergence issue due to large number of soil restraints, the option of removing the friction at several non-convergent supports would lead to inaccuracy of the results. To overcome this issue, stress model with a larger support spacing (<50D Typical) may be utilized. In this case, the vertical support stiffness to be used shall be determined using the above mentioned procedure for finding the soil support stiffness with the increased support spacing. In alternate method the representative vertical stiffness values as per Geotechnical report, after consulting the Civil, can be used to simulate the soil springs for surface laid lines.\n\n\n\nTo obtain the near realistic results, it is suggested to model with actual profile of the flowline so as to gain the proper axial and lateral displacements.\n\n\n\nThe above article gives a broader perspective of general considerations to be taken care of.