Critical Defect Length (CDL)

From AS2885 INFO
Jump to navigation Jump to search
References:
AS2885.1 Clause 5.5.4 Critical defect length
AS2885.1 Clause 4.9 Provisions for high consequence areas 
AS2885.1 Clause 4.9.2 No rupture

"No Rupture Pipe": for new pipelines you design for 1.5xCDL - OK. But for assessing existing pipelines, shouldn't having minimum hole size > CDL satisfy no rupture?

There isn't any difference between existing and new pipelines in this context. The 1.5 factor is a safety margin reflecting uncertainty in the CDL calculation as well as uncertainty in other data such as the actual tooth size (because there is not a perfect correlation between machine weight and tooth dimensions). (Peter Tuft)


If the equivalent hole diameter was previously used for defect length assessment rather than the base length, should the pipeline be retrospectively assessed?

I'd suggest you re-do the calculations, and then use those calculations to make a judgement call if you want to wait for the next scheduled SMS or re-assess immediately. (James Czornohalan)


Section 5.5.4 now notes some limits to the application and validation of equations 5.5.4(1) to 5.5.4(6) – a key one being the Folias factor expression is valid only for CDL < D/2. No guidance is provided by AS2885 on what to do if the calculated CDL is > D/2.

There is no unique equation for Folias factor which is also known as Surface Correction or Bulging Factor and they have all evolved in the past 30-40 years. Almost all of the available equations have been developed based on simple or complex analytical solutions, Finite Element Analysis and some validated through experiments.

The original equation in AGA NG-18 report is one of the early versions which at the time, was believed to have the validity range dependent on defect length and pipe diameter only (L/D or c/R). The limit shown on AS 2885.1-2018 (CDL<D/2) is also a legacy one based on this original equation.

It was later found that validity range of the equation depends not only on the defect length and pipe diameter, but also pipe thickness. This led to the revised equation which was used in ASME B31G. Although the equation presented in ASME B31G looks different at the first glance, it is the same as original AGA NG-18 but written around diameter and defect length rather than radius and half defect length.

In ASME B31G, for the situations where z is larger than 50, a new simplified equation was proposed in addition to the original one. The combination of two presented equations provide a complete solution. These equations are widely used in the pipeline integrity management/assessment.

Please note that Folias factor was initially developed for through-wall defects and later was expanded for part-through defects. Even those equation which present Folias factor for part-through cracks are usually valid where the defect depth is less than 80% of the pipe thickness (d/t<=0.8) and standards such as API 579-1 require that deeper defects to be treated as through-wall defect with adjusted length.

Additionally, if the calculated CDL exceeds D/2, the results need to be reviewed against the credible defect sizes that can be created. The usual CDL calculations are done based on the assumption that material is a high toughness i.e failure is purely due to Plastic Collapse and Toughness independent. Therefore, if a material does not qualify as high toughness material, the CDL needs be calculated from different equations (see AGA NG-18) which take Toughness into consideration.

Usually, the required CVN is calculated first and then the corresponding CDL. Given that CVN can become very high, the corresponding CDL can be high as well. There are situations that with high toughness material and/or thick pipe, the CDL will be longer than the pipe Diameter whilst any defect length above 50-60% of pipe diameter can be technically considered as a full rupture. In those cases, when the credible threats (e.g. Excavator etc.) are assessed, it is the maximum expected defect length of the largest threat that should be considered and not necessarily CDL which is a purely theoretical value. (Mehdi Fardi)