This article describes the uniaxial collapse, burst, axial and compression design factors, and the triaxial design factor, recommended for use, together with a brief overview of the considerations which led to the choice of the generally accepted values.
Each uniaxial design factor is defined as the minimum ratio required between the corresponding casing strength tabulated in API 5C2, on the basis of the formulae of API 5C3 (corrected to take into account the effects of corrosion, wear and fatigue) and the estimated design load. The triaxial design factor is defined as the minimum ratio required between the yield strength (similarly corrected for the effects of corrosion, wear and fatigue) and the Von Mises equivalent stress.
These design factors are "combined" design factors, taking into account both the uncertainties in the manufacturing process leading to variations in casing strength, and those in the design-load estimation process.
Such a combined design factor should not be confused with a safety factor, which is a multiplier to be applied to the maximum design load. The former is based on scientific considerations, while the latter is usually arbitrarily chosen to give a certain resiliency to the design. For casing design, this safety factor should be set equal to unity.
The values used in the drilling industry vary quite widely between operators, because of variations in the design method used. Some include wear or wall-thickness tolerances in the design factors for casing strength. Some operators assume full evacuation to calculate the design load for collapse, while others apply a partial evacuation rule.
Tighter controls in the pipe-manufacturing process have led to an improvement in metallurgical and dimensional properties and hence to more accurately defined casing strengths. This might suggest that the design factors could be reduced. However, there are very few data available to rely on. A probabilistic evaluation of the existing design code would be required to make technically justified changes in the value of the relevant design factors. Quantitative Risk Assessment (QRA) can also assist the selection of design factors.
A summary of the applicable design factors is given below:
- Uniaxial collapse design factor1.0
- Uniaxial burst design factor1.1
- Uniaxial tension design factor1.3
- Uniaxial compression design factor1.0
- Triaxial design factor1.25
Collapse design factor
Using a partial evacuation Design Load case rather than a full evacuation one is considered more realistic. Also the behaviour of cement and annular fluids with time should be considered in the Design Load cases.
The reliability and characteristics of casing collapse capacities is high as a function of the more tightly controlled manufacturing processes. API 5C2 presents the relevant values for the collapse capacity of the casing, but recent studies reveal that these values are occasionally conservative.
Based on the above, and since corrosion, wear and downrating because of tension and temperature should be treated separately, the uniaxial Collapse Design Factor of 1.0 is recommended for collapse design.
Burst design factor
The burst Design Loads have not significantly changed, except for the introduction of cement and annular fluid behaviour. However, the expected Design Loads from possible field scenarios can be refined due to the accumulation of experience and application of prediction techniques.as discussed in the article on Design Parameters.
As is documented in API 5C3, the burst capacity of a casing is related to the yield strength of the material. Hence a conservatism is built into the values as tabled in API 5C2, since initial yielding in burst loading will not rupture the pipe. However, the whole of the casing design is predicted upon the avoidance of yielding. Therefore, no allowance should be made for the fact that rupture of the casing is unlikely even if the burst rating is slightly exceeded. Also, when evaluating the burst capacity of a casing, a down rating because of wear, corrosion, temperature and applied compression is required before the design factor is introduced.
While a collapse failure would normally be expected deeper down the wellbore, the rupture of a casing will most likely be a near-surface event. Hence, the consequences are more severe for such a failure. Based on this consideration, although the probability of the failure mode is low, it is recommended that an uniaxial Burst Design Factor of 1.1 is kept for burst design.
Tension design factor
Software applications can assist in the tension load prediction during the installation phase as well as during the service life time, taking into consideraton the pressure (buoyancy) load, the bending load, the dynamic loads like drag and shock loads, and the changes in axial load by changes in temperature and pressurese. However, it should be highlighted that the static drag loads are more difficult to quantify.
The uncorrected value for the tension capacity of a casing string is presented in API 5C2. However, when evaluating the tension capacity of a casing a down rating because of wear, corrosion and temperature is required before the Tension Design Factor is applied.
Based on these considerations, it is recommended that an uniaxial Tension Design Factor of 1.3 is kept in casing design.
Compression design factor
It has been demonstrated that casing failure due to compressive loading will be mainly a result of elastic or plastic instability, i.e. helical buckling. A pure compression failure, i.e. casing squashing, is unlikely in most cases.
The casing resistance against buckling can be significantly increased by the placement of centralisers. If the relevant casing is rigidly supported by centralisers, very high compression loads can be carried before buckling occurs.
It can be seen that two Design Factors result:
·If buckling is not possible because of the placement of centralisers between the casing string under consideration and the previous casing string it is recommended that a uniaxial Compression Design Factor of 1.0 is used. This can be justified because the calculated centraliser spacing inherently covers a buckling Design Factor of 1.5.
·If buckling is acceptable a post-buckling analysis should be carried out to establish the relevant triaxial stress state. In analogy with the discussion on the triaxial Design Factor below, the same Triaxial Design Factor of 1.25 is recommended for these situations.
Triaxial design factor
Translation of the load conditions into a three-dimensional stress state is possible with design softwares. The refinement of the Design Load cases and the Design Parameter, due to the accumulation of experience and prediction techniques, leads to a smaller distribution in the Design Loads. Wear and corrosion related wall thickness reductions should be taken into account in this stage. As a result the three-dimensional stress state is more realistic.
A comparison of this resulting three-dimensional stress state with the yield strength value (corrected for temperature) of the uniaxial test is commonly achieved via the von Mises yield criterion. This yield criterion has been extensively used and repeatedly verified. The direct comparison of this von Mises equivalent stress to the yield strength of the material provides a single design factor.
Based on field experience with triaxial analyses in tubing design and the analogy with casing design, it is recommended that a Triaxial Design Factor of 1.25 is used for casing design analyses.